Effects of taxol on choriocarcinoma cells

Effects of Taxol on choriocarcinoma cells Christian Marth, MD, Thomas Lang, MD, Martin Widschwendter, MD, Elisabeth Mtiller-Holzner, MD, and Gtinter Daxenbichler, PhD Innsbru&, Austria OBJECTIVE: Taxol (Bristol-Myers Squibb) (paclitaxel) has been shown to be a potent inhibitor of cell growth for a variety of tumors. We were interested in the antiproliferative efficacy and biologic properties of this novel antineoplastic agent in choriocarcinoma cells. STUDY DESIGN: Human choriocarcinoma cell lines JAR and BeWo were cultured as monolayers and treated with Taxol. RESULTS: Proliferation of JAR and BeWo cells was inhibited by Taxol in a dose-related manner and 1 to 3 nmol/L was sufficient to achieve 50% growth reduction. This effect was accompanied by a marked induction of human chodonic gonadotropin secretion, The effect on human chorionic gonadotropin secretion was dependent on intact protein biosynthesis but not mediated by augmented messenger ribonucleic acid expression. In these choriocarcinoma cells Taxol promoted differentiation as shown by an increase in syncytiotrophoblastic-like cells. Combination of Taxol with either etoposide or methotrexate resulted in antagonistic growth inhibition. CONCLUSION: Taxol is a highly effective antineoplastic agent in choriocarcinoma cells, and clinical trials in refractory disease would therefore be warranted. However, substanoes other than etoposide or methotrexate should be evaluated for combined treatment. In addition to growth inhibition, differentiation is also induced by Taxol, as shown by increased human chorionic gonadotropin secretion and changed morphologic features. (AM J OBSTETGYNECOL1995; 173:1835-42.)

Key words: Taxol, choriocarcinoma, differentiation, human chorionic gonadotropin Taxol (Bristol-Myers Squibb, Princeton, N.J.) (paclitaxel), a chemotherapeutic agent derived from the bark of the Pacific yew tree, has been shown to exert promising activity in a variety of cancer cells in culture.' High in vivo efficacy was also confirmed in phase II trials, in particular for breast and ovarian cancer patientsY' :~ Unusually high response rates were reported for cisplatin-refractory ovarian cancer. ~ Efficacy of Taxol in heavily pretreated patients could be explained by its unique mechanism of action. Taxol inhibits cell division by promoting the assembly of microtubules and stabilizing tubulin polymers by preventing their depolymerization.L 5. 6 Taxol-treated cells exhibit characteristic bundles of microtubules that are not usually associated with the microtubule organizing center, the centrosome. Mitosis is inhibited because the two centrosomes, which form the two spindle poles during metaphase, do not have necessary microtubules associated with them. Therefore mitotic cells (M phase) are sensitive for Taxol, whereas interphase cells (Gl, S, and G 2 phase) are resistant. 7

From the Department of Obstetrics and Gynecology, lnnsbruck University Hospital. Received for publication September 7, 1994; revised FebruaU 27, 1995; accepted April 12, 1995. Reprint requests: Christian Marth, MD, Univ. Klinih far Frauenheilkunde, Anichstra~Se 35, A-6020 Innsbruck, Austria. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/65586

'Although several tumor types have been evaluated, to our knowledge no information is available about the action of Taxol in choriocarcinoma. "Gestational trophoblastic disease," including its malignant form (namely, choriocarcinoma), is a term that describes a continuum of tumors arising in the fetal chorion of the placenta. Currently four primary types of gestational trophoblastic disease are recognized, which can be differentiated on the basis of histopathologic and cytogenetic findings and clinical features. ~ They are complete hydatiform mole, partial hydatiform mole, gestational choriocarcinoma, and placental site trophoblastic tumor. The International Federation of Gynecology and Obstetrics proposed a staging system for gestational trophoblastic disease based on the location of the tumor: stage I, corpus; stage II, outside the uterus but limited to genital structures; stage III, lung metastases; and stage IV, other metastatic sites. A classification scheme which allows the estimation of prognosis was proposed by Newlands et al. and in a simplified manner by the World Health Organization. 9 This scoring system divides patients into three risk categories (low, medium, and high) and allows selection of treatment on the basis of these assignments. The usefulness of this classification has been demonstrated by evidence that all patients in the medium-risk group can be cured with combination chemotherapy. But in those patients with multiple poor risk factors treatment failures may occur. In this group, estimated to include about 15% of the 1835

~836 Marth et al.

population with metastatic gestational trophoblastic disease, the regimen of etoposide, methotrexate, actinomycin D, cyclophosphamide, and vincristine sulfate has been shown to be effective and to induce a cure rate of about 80%. 8, 10. 11 There remains, however, a small group who will also not respond to salvage therapy, including cisplatin. IntrodUction of new active substances lacking cross-resistance with other chemotherapeutic agents would therefore further improve survival rate. The aim of this study was to evaluate a potential role of Taxol in the treatment of choriocarcinoma. For this purpose we used two human choriocarcinoma-derived cell lines, namely, JAR and BeWo, which were cultured as monolayers. In gestational troph0blastic disease determination of serum human chorionic gonadotropin (hCG) concentration is highly sensitive and specific for diagnosis and foll0w-up. However, several biologic response modifiers and chemotherapeutic agents interfere with the synthesis and release of hCG by choriocarcinoma cells. ~2 Induction of hCG production by, among others, methotrexate or etoposide has been considered a sign of differentiation from cells with the characteristics of the cytotrophoblast with those with the characteristics of the syncytiotrophoblast/s' J4 We were therefore interested in whether Taxol is able to increase the synthesis of hCG and induce differentiation in these cells. Material and methods Substances. Taxol and etoposide were kindly provided by Bristol-Myers Squibb, and methotrexate was obtained from Lederle (Vienna, Austria). Cycloheximide was purchased from Sigma (St. Louis). Cell culture. T h e h u m a n choriocarcinoma cell lines JAR and BeWo were kindly provided by Dr. P. Berger, Department of General and Experimental Pathology, University of Innsbruck (Austria), and cultured under standard conditions in minimum essential m e d i u m supplemented with 10% fetal bovine serum (both from' Biological Industries, Kibbutz Beth Haemek, Israel), as recently described. 12 The BeWo cell line, the first human endocrine cell type to be maintained in continuous cultivation, was initiated in 1966 from a cerebral metastasis of a choriocarcinoma. ~ JAR cells were established directly from a trophoblastic tumor of the placenta, ~6 and produce hCG, estrogen, and progesterone similarly to BeWo cells. To compare the efficacy of Taxol in ch0riocarcinoma cells and in cell lines from Taxol-sensitive tumors, six ovarian carcinoma cell lines were investigated: SKOV-6, HTB-77, OVCAR-3, HOC-7, 2774, and 2780. Cells were kindly provided by Dr. G. Gastl, Memorial Sloan Kettering Cancer Center, New York, and Dr. C. Dittrich, University of Vienna,

December 1995 Am J Obstet Gynecol

and cultured under standard conditions. Culture medium used for ovarian carcinoma cells was the same as for choriocarcinoma. For determination of growth effects, carcinoma cells were seeded in 24-well t i s s u e culture plates (Nunc 146485, Roskilde, Denmark) and allowed to attach overnight. Taxol or the other chemotherapeutic agents were a d d e d in the desired concentrations (indicated in legends to tables and figures). All substances applied remained in the culture m e d i u m for the entire incubation period, after which supernatant medium was collected for hCG determination. The continuous treatment was chosen in agreement with the recommendations of Lopes et al., v who showed that cell kill is related more to increased exposure time than to increased drug concentration. This behavior is explained by the phase-specific action of Taxol: increasing the treatment period may increase the fraction Of cells entering a Taxol-sensitive phase of the cell cycle. The specimens for hCG determination were centrifuged for 10 minutes at 10,000g to remove debris and stored at - 7 0 ° C until analysis. After dilution with serum diluent (1:5) hCG concentration was determined with an immunoradiometric assay specific for total hCG and its 13-chain (HCG-MAIA-CLONE, Serono, Durrey, United Kingdom). Measured hCG concentrations were corrected for the number of cells. Results were either expressed as units per 100,000 cells or corrected for the number of cells in each group in relation to untreated controls at each time point and expressed as units p e r liter. Moreover, human placental lactogen was determined in cell culture supernatant from control and Tax01-treated JAR and BeWo cells by means of a specific immunoradiometric assay (Amerlex h u m a n placental lactogen, Kodak Clinical Diagnostics, Amersham, United Kingdom). Cells were detached from tissue culture plates with the help of trypsin (0.05%)-ethylenediaminetetraacetic acid (0.02%) in Dulbecco's phosphate-buffered saline solution (Biological Industries), and their number was assessed with an electronic particle counter (Coulter, Dunstable, United Kingdom). Viability of cells was checked by trypan blue exclusion and found to be >95% in each experiment for controls and Taxoltreated cells. Moreover, in the suspension p r e p a r e d for the Coulter counter, the number of cell clusters was always > 4%. To investigate whether Taxol-mediated hCG induction is protein biosynthesis dependent, choriocarcinoma cells were t r e a t e d simultaneously with cycloheximide (5 Ixg/ml)and Taxol. Differentiation. To evaluate whether Taxol induces differentiation, JAR and BeWo cells were cultured on glass slides and treated for 5 days with or without 3 nmol/L Taxol. After fixation with acetone in phosphate-buffered saline solution, hematoxylin-eosin stain-

Volume 173, Number 6 Am J Obstet Gynecol

Marth et al.

100

1837

~1oo

I

500

'.,..,.,_ 0"~ oI,,= o o o

::3"

o ....

-%,...

0

• •, ......

"~ o¢.) 50

%

400 _ 0

"... ;:

3O

30 "~ 0

o

'~

o

300 o o_

o-o 20

0-0

JAR BeWo

e

!

200

:

i [[~

10

o .s

..-" 8"

E C

10

i (~.• "•%

51

, II .... o," control .I

, .3 taxol

I

control

I

I

I

I

I

1

3

10

30

100

taxol

(nM}

Fig. 1. Effects of Taxol on proliferation of choriocarcinoma (. . . . ) and ovarian carcinoma ( - - ) cells. Human choriocarcinoma cell lines JAR (C)) and BeWo (e) and human ovarian cancer cell lines SKOV-6 (z~), HTB-77 (e), OVCAR-3 (I), HOC-7 (O), 2774 ([]), and 2780 (A) were continuously treated with Taxol for 5 days in concentrations indicated on abscissa. Results are presented as mean number of six wells counted in percentage of untreated control. Coefficient of variation was always < 10% and is not shown for reasons of clarity.

ing was p e r f o r m e d . M o r p h o l o g i c features and classification into cytotrophoblastic and syncytiotrophoblastic cells were d e t e r m i n e d according to the criteria of Friedm a n and Skehan.'4 Northern blot analysis. Ribonucleic acid (RNA) p r e p a r a t i o n and analysis were p e r f o r m e d as recently described. '7 Briefly, cells r e a c h i n g confluency were treated with or without T a x o l (3 nmol/L) for 48 hours. T o t a l R N A was p r e p a r e d with the g u a n i d i n e thiocyanate m e t h o d . R N A was resolved on a 1% agarosef o r m a l d e h y d e gel, transferred to nylon m e m b r a n e s , ultraviolet cross-linked, and hybridized in a f o r m a m i d e based hybridization solution at 56 ° C with a digoxigenin-labeled 41 bp c o m p l e m e n t a r y deoxyribonucleic acid f r a g m e n t of the h C G 131 gene. Is Detection of d i g o x i g e n i n - l a b e l e d nucleic acids by c h e m i l u m i n e s cence e n z y m e i m m u n o a s s a y on nylon m e m b r a n e s was carried out following m a n u f a c t u r e r ' s instructions (DIG L u m i n e s c e n t Detection Kit, B o e h r i n g e r M a n n h e i m Biochemica, Vienna, Austria). Statistics. Data were analyzed with the use of non-

.'"'"

e..............e.:.......,,..]o .......... o o '

•-

.2.

•

, I

100

{nM)

Fig. 2. Effects of Taxol on proliferation (----) and hCG secretion ( . . . . ) of JAR and BeWo choriocarcinoma cells• Cells were treated for 5 days as indicated on abscissa; cell number was counted and hCG concentration determined as described in Material and Methods• hCG concentration was corrected for cell number and related to untreated controls. Cell numbers are given as mean of six wells counted in percentage of untreated control• Coefficient of variation was always < 10% and is not shown for reasons of clarity.

p a r a m e t r i c tests such as the Wilcoxon U test. T h e n m h i p l e d r u g effect analysis, which is based on the median-effect principle, was used to calculate c o m b i n e d d r u g effects of T a x o l with either m e t h o t r e x a t e or etoposide according to r e c o m m e n d a t i o n s of B e r e n b a u m x~ and C h o u and Talalay ~° by applying the c o m p u t e r p r o g r a m "Dose effect analysis with m i c r o c o m p u t e r " by C h o u and C h o u (Elsevier Biosoftware, Cambridge, U n i t e d Kingdom). This m e t h o d involves plotting of dose-effect curves for each a g e n t and their c o m b i n a tions in multiply diluted concentrations by use of the median-effect equation:

fa/fu = (D/Om)"

(1)

W h e r e D is the dose, Dm is the dose r e q u i r e d for 50% inhibition of cell growth, fa is the fraction of cells affected by dose D, fu is the unaffected fraction, and m is a coefficient of the sigmoidicity of the dose-effect curve; m = 1, m > 1, and m < 1 indicate hyperbolic, sigmoidal, and negative sigmoidal dose-effect curves, respectively, for an inhibitory drug. Dose-effect curves were plotted with either a g e n t alone a n d with their combination. F r o m these e x p e r i m e n t s a c o m b i n a t i o n index was calculated to characterize quality of interaction:

1838

Marth et al.

December 1995 Am J Obstet Gynecol

500

400

0 0 400 300

300 o. (9

u

o

200

200

o 100 100

,

,

,

,

1

2

3

4

0

,

,

',

1

2

days

'

,

T

3

4

,

days

Fig. 3. Time course of hCG secretion (right panel) and proliferation (left panel) ofTaxol-treated BeWo cells. Choriocarcinoma cells were cultured in presence of Taxol (O control; e, 1 nmol/L; [Z, 1.8 nmol/L; # , 3 nmol/L) for up to 5 days. Cell number and hCG were determined as described in Material and Methods. hCG concentration was normalized to cell number of controls of each time point. Coefficient of variation was always < 10% and is not shown for reasons of clarity.

10OO

300

BeWo "~

-~

JAR

800

~

250

600

i 200

O

~,

?,00 .l=

0

,

~-

control

CHX

~

taxol taxol+OHX

0

~-

control

CHX

taxol ta×ol+CHX

Fig. 4. Effects of cycloheximide (CHX) on Taxol-mediated hCG induction in choriocarcinoma cells. BeWo (left panel) and JAR (right panel) cells were treated as indicated on abscissa without additives (control), cycloheximide (5 txg/ml), Taxol (hatched box, 1 nmol/L; solid box, 1.8 nmol/L; open box, 3 nmol/L) alone or in combination with cycloheximide (5 ixg/ml). Results are expressed as mean value of four wells measured + 1 SD.

T a b l e I. Effects o f T a x o l o n d i f f e r e n t i a t i o n o f c h o r i o c a r c i n o m a cells

Cell line

Control

Taxol (3 nmol/L)

Significance

JAR BeWo

6% -+ 5% 8% -+ 4%

25% -+ 7% 36% + 8%

p < 0.05 p < 0.05

Results are given as mean percentage of syncytiotrophoblastic cells -+ 1 SD.

C o m b i n a t i o n i n d e x = (D)l/(Dx)l + (D)2/Dx)~ (2) E q u a t i o n 2 dictates t h a t d r u g 1 (i.e., [D]I a n d d r u g 2 (i.e., [D]2 ) in t h e n u m e r a t o r s in c o m b i n a t i o n i n h i b i t x%. (D~)1 a n d (D~),~ in t h e d e n o m i n a t o r s are t h e doses o f d r u g 1 a n d d r u g 2 alone, respectively, t h a t also i n h i b i t x%. C o m b i n a t i o n i n d e x - 1 i n d i c a t e s a n a d d i t i v e inter-

action, w h e r e a s c o m b i n a t i o n i n d e x > 1 r e p r e s e n t s antagonistic a n d c o m b i n a t i o n i n d e x < 1 synergistic activity. T h i s type of c a l c u l a t i o n is very' similar to isobolog r a m construction a n d has t h e a d v a n t a g e to p r e v e n t e r r o r s c a u s e d by d o s e - r e s p o n s e curves ' t h a t are n o t simply e x p o n e n t i a l . ~*~'~o

Volunle 173, Number 6 .am J Obstet Gynecol

Marth et al.

,~~

etoposide (ng/ml, a-a) 0.0 100 I L l - i t

5

7.5

10

12.5

i

i

L

I

methotrexate

15

1839

o.o

4 --.--i1._. =

lOO

(nM, o - o )

6 r

8 =

10 i

115

23

215

12

75 ,n

*o 50-

"

~

50-

E g 25-

25-

JAR

JAR 0

o.o

LI

A

1'.o ,'.5 2'.o 2'.5 taxol (nM, o-o)

etoposide 0,0 100.~i

I

(ng/ml,

0

3.0

O.0

37.5

50

62.5

I

I

I

I

1:0

methotrexate 75

3.0

(nM, o - o ]

taxol

r•-n)

25

p,

(nM, n - n )

0.0

5

7.5

10

~1

L

=

r

12.5

15

1O0 - ~

75-

75-

__=

o

!

,n o

50-

50-

25-

25-

BeWo 0 i 0,0

B

II

i 1.0

, 1.,5 taxol

i 2.0

i 2.5

0-~ 3.0

(riM, o - o 1

0.0

110

115

taxol

210

215

3.0

(nM, o - o ]

Fig. 5. Effects of etoposide, methotrexate, and Taxol alone (open symbols as indicated in chart) or combination of etoposide with Taxol or methotrexate with Taxol (e - • in respective chart) on proliferation of JAR and BeWo cells. Number of cells was determined and experiments were performed as described in Material and Methods to allow application of the mathematic model of Chou and Talalay.

Results

Proliferation of JAR and BeWo ceils was inhibited by Taxol in a dose-dependent m a n n e r (Figs. 1 and 2). The concentration of Taxol producing 50% growth inhibition range between 1 and 3 nmol/L. The six ovarian carcinoma cell lines tested were also affected by Taxol; but the 50% growth inhibition ranged between 3 and 10 nmol/L (Fig. 1). On the other hand, concentrations of hCG in supernatant medium of choriocarcinoma cells increased as the cell n u m b e r decreased (Fig. 2). At 3 nmol/L Taxol a threefold to fivefold augmentation of hCG production per cell was measured. Action of Taxol on both proliferation inhibition and induction of hCG reached statistical significance (p < 0.05) on day 2 (for 3 nmol/L Taxol only) and m a x i m u m effectiveness on

day 4 (for each dose tested, p < 0.05, Fig. 3). This action was similar for both cell lines, although the amount of hCG secreted was about five times higher for BeWo than for JAR cells. In contrast to the effects on hCG, we were unable to measure h u m a n placental lactogen in supernatants obtained from control or Taxol-treated JAR or BeWo cells (data not shown). To evaluate whether Taxol-induced hCG secretion is protein biosynthesis dependent, cells were treated with cycloheximide (5 ixg/m!). Incubation with cyeloheximide alone completely abolished hCG release of both cell lines, and in combination with Taxol it was able to effectively reduce the Taxol-mediated hCG induction in JAR and BeWo cells (Fig. 4). Because the mechanism of hCG induction could be

1840

December 1995 Am J Obstet Gynecol

Marth et al.

5

choriocarcinoma patients the two most important substances. To look into possible clinically active regimen, we were interested in testing the activity of these two agents in combination with Taxol in vitro. Proliferation of JAR and BeWo cells was inhibited by methotrexate and etoposide in a dose-related manner (Fig. 5). A combination of Taxol with either methotrexate or etoposide resulted in increased antiproliferative activity, but its growth inhibition was only slightly improved c o m p a r e d with the single agents. Analysis of interaction by calculation of the combination index dearly revealed an antagonistic interaction of both agents and for both cell lines (Fig. 6).

4

g ¢-

o

Comment 1

'

0 0.0

0.2

0.4

0.6

0.8

1.0

fraction affected Fig. 6. Computed plot of combination index with respect to inhibition of proliferation by etoposide plus Taxol (C) - @, JAR; • - •, BeWo cells), methotrexate plus Taxol ([Z x [7, JAR; • - •,, BeWo cells). Combination index < 1, combination index = 1, and combination index > 1 indicate synergism, additivity, and antagonism, respectively. Inhibition of proliferation is expressed as fraction affected; fraction affected 0.7 indicates, for example, 70% inhibition compared with control group. Calculation was performed as described in Material and Methods according to Chou and Talalay by use of data from Fig. 5. Computer simulated combination index values at various effect levels to generate combination index plots. These combination index values were found at all effect levels (fraction affected) > 1. This would indicate antagonism for each pair of combination regardless of dose applied or fraction of cells affected.

mediated at either a transcriptional or posttranscriptional level, we analyzed specific hCG RNA. In JAR cells hCG messenger RNA was expressed at a very low level, whereas BeWo cells showed high expression levels. Neither in JAR nor BeWo cells did treatment with Taxol at a concentration of 3 nmol/L for 48 hours have any effect on hCG messenger RNA expression level (data not shown). JAR and BeWo cells grown on slides with and without Taxol (3 nmol/L) treatment were analyzed (200 cells per group and three i n d e p e n d e n t experiments) and classified as cytotrophoblastic or syncytiotrophoblastic by a pathologist not informed of the treatment. After a 5-day treatment with Taxol, the median percentage of syncytiotrophoblastic cells increased significantly for both JAR and BeWo cells (Table I). Methotrexate and etoposide are for treatment of

This is the first demonstration of the high antiproliferative efficacy of Taxol in JAR and BeWo choriocarcinoma cells. Ta:~ol was found to be about three times more effective in these gestational trophoblastic tumor cells than in six ovarian carcinoma cell lines. Values for 50% growth inhibition determined in this study for the ovarian carcinoma cell lines (3 to 10 nmol/L) were higher than those published by Kelland and Abel, '-'1who with a mean 50% growth inhibition of 1.1 nmol/L Taxol observed in six ovarian carcinoma cell lines had the best results in literature. Direct comparison of 50% growth inhibition values from various studies is questionable, however, because differences in methods applied may markedly affect results. For example, Kelland and Abel 2t used the sulforhodamine B assay, which is a colorimetric procedure and may primarily estimate cytotoxicity, whereas we determined proliferation by counting cell number. These differences in method could be one explanation for divergent results. We preferred to count cells for estimation of cell growth because this method has the advantage of measuring the number of surviving cells after a long time of drug exposure and allows observation also of the secondary effects induced by the applied agent. In other studies 50% growth inhibition values reported for Taxol in ovarian carcinoma cells (4 or 5 nmol/L in the case of A2780 or OVG1 cells, respectively) were clearly higher and in the range of our observationT' = 50% Growth inhibition values obtained for choriocarcinoma (1 to 3 nmol/L) thus indicate at least a similar (compared with Kelland and Abel) or higher (compared with H a h n et al. or Lopes et al.) sensitivity for Taxol. In any case our data would suggest high sensitivity of choriocarcinoma cells for this novel antineoplastic agent, and clinical trials in chemotherapy-refractory patients seem advised. It should be noted that plasma concentrations of Taxol in treated patients are substantially higher than the 50% growth inhibition values obtained in the current study. ~a However, the two most important therapeutic drugs for gestational trophoblastic disease, namely, methotrexate and etoposide, seem not to be

Volume 173, Number 6 ,,M'n J Obstet Gynecol

preferred candidates for combination therapy with Taxol because antagonism has been observed for both substances and both cell lines. Antagonism has been defined as interaction of the two drugs less than would expected when the activities of both substances are added. For the combination of Taxol with either etoposide or methotrexate more choriocarcinoma cells were killed than by any single agent alone but less than would be predicted by a mathematic method assuming no interaction. This was confirmed by the curves in Fig. 6 constructed by means of the median-effect equation that applies the principle of the mass-action law. For the entire dose range studied the combination index was greater than one indicating antagonistic interaction. Antagonistic growth inhibition of a combination of etoposide with Taxol has also been described for breast cancer cells. An explanation for this unfavorable effect could be, as mentioned by H a h n et al., 2~ the cell cycle arrest in G2/M phase provoked by Taxol. The two other agents primarily affect processes necessary to pass through the S phase: methotrexate inhibits dihydrofolate reductase, an enzyme needed to provide the cell with nucleotides, and etoposide reduces topoisomerase II activity, an enzyme involved directly in deoxyribonucleic acid replication. These two chemotherapeutic agents would thereby reduce the fraction of cells entering a Taxol-sensitive phase of the cell cycle. Combination of Taxol with other chemotherapeutic agents should therefore be evaluated for possible treatment of choriocarcinoma patients. The substances of choice would be those with a mechanism of action not restricted to one specific phase of the cell cycle. Another possibility could be a careful sequencing of drug treatment to avoid such antagonistic actions, as described for cisplatin and Taxol. 24 This would explain the high remission rates for a combination of Taxol with cisplatin which have recently been reported for treatment of ovarian cancer patients? In addition to growth inhibition, Taxol augmented in a dose-related m a n n e r the secretion of hCG in both cell lines. This effect was d e p e n d e n t on intact protein biosynthesis, as shown by the experiments with cycloheximide, but not regulated at a transcriptional level, as shown by Northern blot analysis. It might be argued that the increased release of hCG observed in supernatant medium could be Taxol-induced cell lysis. However, by trypan blue exclusion we were unable to show an increased cell lysis in Taxol-treated cells. Moreover, incubation with cycloheximide augmented antiproliferative activity (data not shown) but eliminated hCG release. O n the other hand, morphologic analysis clearly revealed that Taxol induces differentiation in both choriocarcinoma cell lines. This was demonstrated by an increased percentage of cells with a syncytiotrophoblastic phenotype. The multinuclear choriocarcinoma cells observed may arise from either cell fusion

Marth et al.

1841

(as was shown for placental cytotrophoblasts) or karyokinesis not followed by cytokinesis. The latter possibility would fit well with what is known about the mechanism of Taxol action. Stabilization of microtubuli as induced by Taxol results in the inhibition of mitosis and interphase functions such as maintenance of cell shape, cell motility, and intracellular transport but not in complete inhibition of deoxyribonucleic acid replication. Other agents that induce differentiation in choriocarcinoma cells, like methotrexate or hydroxyurea, are known to augment hCG messenger RNA. In contrast to these substances, we were unable to detect any effect on Taxol treatment at the transcriptional level. One explanation could be a posttranscriptional regulation of hCG biosynthesis by Taxol, which may not be reflected by a change in RNA concentration. We thank Mrs. J. Rbssler for her skillful technical assistance.

REFERENCES

1. Rowinsky EK, Cazenave LA, Donehower RC. Taxoh a novel investigational antimicrotubule agent. J Natl Cancer Inst 1990;82:1247-59. 2. Holmes FA, Waiters RS, Theriault RL, et al. Phase II trial of taxol: an active drug in the treatment of metastatic breast cancer. J Natl Cancer Inst 1991;83:179%805. 3. McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Look KY, Partridge. A phase III trial comparing cisplatin/Cytoxan and cisplatin/Taxol in advanced ovarian cancer. Proc Am Soc Clin Oncol 1993;12:255. 4. Thigpen T, Blessing J, Ball H, et al. Phase II trial of Taxol as second-line therapy for ovarian carcinoma: a Gynecologic Oncology Group Study. Proc Am Soc Clin Oncol 1990;9:604. 5. Schiff PB, Fant J, Horwitz SB. Promotion of microtubule assembly in vitro by Taxol. Nature 1979;277:665-7. 6. Schiff PB, Horwitz SB. Taxol assembles tubulin in the absence of exogenous guanosine 5'-triphosphate or microtubule associated proteins. Biochemistry 1981;20:324751. 7. Lopes NM, Adams EG, Pitts TW, Bhuyan BK. Cell kill kinetics and cell cycle effects of taxol on human and hamster ovarian cell lines. Cancer Chemother Pharmacol 1993;32:235-42. 8. Lewis JL Jr. Diagnosis and management of gestational trophoblastic disease. Cancer 1993;71:1639-47. 9. Newlands ES, Bagshawe KD, Begent RHJ, Rustin GJS, Holden L. Results with the EMA/CO (etoposide, methotrexate, actinomycin D, cyclophosphamide, vincristine) regimen in high risk gestational trophoblastic tumours, 1987 to 1989. BrJ Obstet Gynaecol 1991;98:550-7. 10. Rustin GJS, Bagshawe KD, Hammond CB. Gestational trophoblastic tumors. CRC Crit Rev Oncol/Hematol 1985; 3:103-42. 1I. Smith DB, O'Reilly SM, Newlands ES. Current approaches to diagnosis and treatment of gestational trophoblastic disease. Curr Opin Obstet Gynecol 1993;5:84-91. 12. Marth C, Brania M, Mtiller-Holzner E, et al. Modulation of secretion of human chorionic gonadotropin by biologic response modifiers on term placenta and choriocarcinoma ceils. Mol Biother 1989;1:140-4. 13. Speeg KV, Azizkhan JC, Stromberg K. The stimulation by methotrexate of human chorionic gonadotropin and placental alkaline phosphatase in cultured choriocarcinoma cells. Cancer Res 1976;36:4570-6. 14. Friedman SJ, Skehan P. Morphological differentiation of

Baggish et al.

15. 16. 17. 18. 19. 20.

human choriocarcinoma cells induced by methotrexate. Cancer Res 1979;39:1960-7. Patillo RA, Gey GO. The establishment of a cell line of human hormone-synthesizing tropoblastic cells in vitro. Cancer Res 1968;28:1231-6. Patillo RA, Ruckert A, Hnssa RO, Bernstein R, Dells E. The JAR cell line: continuous lines of human choriocarcinoma. In Vitro 1971;6:398-9. Chomczynshi P, Sacchi N. Single step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Ann Biochem 1987;162:156-9. Mosaki B, Irving B. Identification of the transcriptionally active genes of the chorionic gonadotropin 13 gene cluster in vivo. J Biol Chem 1992;267:3179-84. Berenbaum MC. Criteria for analyzing interactions between biologically active agents. Adv Cancer Res 1981 ;35: 269-335. Chou TC, Talalay P. Quantitative analysis of dose-effect

December 1995 Am J Obstet Gynecol

21.

22. 23.

24.

relationships: the combined effect of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984;22:27-54. Kelland LR, Abel G. Comparative in vitro cytotoxicity of taxol and taxotere against cisplafin-sensitive and -resistant human ovarian carcinoma cell lines. Cancer Chemother Pharmacol 1992;30:444-50. Hahn SM, Liebmann JE, Cook J, Fisher J, et al. Taxol in combination with doxorubicin or etoposide. Cancer 1993; 72:2705-11. Hnizing MT, Keung ACF, Rosing H, et al. Pharmacokinetics of paclitaxel and metabolites in a randomized comparative study in platinum-pretreated ovarian cancer patients. J CIin Oncol 1993;11:2127-35. Jekunen A, Christen RD, Shalinsky DR, Howell SB. Synergistic interaction between cisplatin and Taxol in human ovarian carcinoma cells in vitro. Br J Cancer 1994;69:299306.

A computer-controlled, continuously circulating, hot irrigating system for endometrial ablation Michael Baggish, MD, Marie Paraiso, MD, E.M. Breznock, DVM, and Stephen Griffey, DVM Cincinnati, Ohio, and Sacramento, California OBJECTIVE: This study evaluated the efficacy of a computer-controlled thermal catheter system to ablate the endometrium by use of 5 to 15 ml of hot saline solution. STUDY DESIGN: Five in vivo sheep uteri and 32 freshly excised human uteri were studied by inserting a self-contained heater catheter equipped with thermocouples and circulating hot saline solution for 15 minutes within the uteri at various temperature settings ranging from 60° C to 80 ° C. The uteri were grossly examined and microscopically evaluated by hematoxylin and eosin and reduced nicotinamide adenine dinucleotide-diaphorase stains for degree of acute destruction, depth of thermal injury, and predicted viability. RESULTS: All 80° C sheep treatments resulted in complete loss of viability at mucosal, submucosal, and myometrial levels. In human uteri temperatures _>80° C produced grossly recognizable thermal damage to the endometrium. Full-thickness irreversible damage was seen in the endometrium. Loss of viability was also demonstrated 1 to 3 mm deep in the superficial myometrium. Intrauterine pressure was demonstrated to range from 20 to 40 mm Hg, with a mean of 26 mm Hg. No leakage of fluid from the oviducts was observed. CONCLUSION: The data presented here demonstrate that a simple technique coupled with a sophisticated monitoring control system could successfully ablate the endometrium without the vagaries associated with individual operator skill levels. (AM J OBSTETGYNECOL1995;173:1842-8.)

Key w o r d s : E n d o m e t r i a l ablation, t h e r m a l catheter, continuously circulating hot saline solution

From the Department of Obstetrics and Gynecology, Good Samaritan Hospital, and the Department of Veterinary Pathology, Davis Veterinary Medical Teaching Hospital, Received for publication December 7, 1994; revised March 31, 1995; accepted April 12, 1995. Reprint requests: Michael S. Baggish, MD, Department of Obstetrics and Gynecology, Good Samaritan Hospital, 375 Dixmyth Ave., Cincinnati, OH 45220-2489. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/65581 1842

Transcervical e n d o m e t r i a l ablation has b e e n shown to be an efficacious and cost-effective operation. In 1981 the first c o n t e m p o r a r y description of this proced u r e using the n e o d y m i u m / y t t r i u m - a l u m i n u m - g a r n e t laser was published.I N u m e r o u s articles have c o n f i r m e d its m a n y benefits? -4 Subsequently, the urologic resectoscope has b e e n a d a p t e d for ablation as well as resection of the e n d o m e t r i u m ? -7 As a g r e a t e r n u m b e r of physi-