John Shepherd Consultant, ?Tim Chard ... placental alkaline phosphatase (PLAP), free p human chorionic gonadotrophin (hCG) and cancer- ..... Devine PL.
British Journal of Obstetrics and Gynaecology September 1997,Vol. 104, pp. 1024-1029
Serum concentrations of cancer antigen 125, placental alkaline phosphatase, cancer-associated serum antigen and free beta human chorionic gonadotrophin as prognostic markers for epithelial ovarian cancer *Thomas Ind Specialist Registrar, ?Ray Iles Lecturer, ?John Shepherd Consultant, ?Tim Chard Professor *Department of Obstetrics and Gynaecology,Hillingdon Hospital, Llxbridge; ?Department of Obstetrics and Gynaecology, St Bartholomewk and The Royal London Hospital School of Medicine and Dentistv, London
Objective To investigate the prognostic significance of elevated levels of cancer antigen 125 (CA125), placental alkaline phosphatase (PLAP),free p human chorionic gonadotrophin (hCG) and cancerassociated serum antigen (CASA) in women with primary epithelial ovarian carcinoma.
Design A two year follow up study of survival. Setting A tertiary care gynaecological oncology unit. Participants One hundred and eleven women with histologically confirmed epithelial ovarian cancer. Main outcome measures Survival over a two year period. Results Stage corrected log-rank x2 tests demonstrated a significant effect on survival for all four tumour markers (CAI25 P = 0.0142; PLAP P < 0.0001; CASA P = 0.0098; hCG P = 0.0002). This was confirmed when each variable was fitted together with disease stage in Cox proportional hazard models. When fitted as multiple variables in a Cox proportional hazard model, the addition of free phCG and CASA to disease stage, PLAP concentrationsand CAI25 levels did not demonstrate M e r prognostic value. Conclusions Levels of all four markers correlate with survival in patients with epithelial ovarian cancer. The combination of PLAP and CA125 concentrations together with disease stage may be used to predict survival but the addition of hCG and CASA levels do not give additional prognostic information.
INTRODUCTION Measurement of circulating tumour markers in cases of epithelial ovarian cancer has been advocated for monitoring disease progression'. More recently their value in predicting prognosis has been studied. In particular, serum concentrations of CA125 and placental alkaline phosphatase (PLAP) have been shown to have an inverse correlation with To date, over 40 monoclonal antibodies against ovarian tumour-associated antigens have been reporteds. It is therefore important to know if serum concentrations of other tumour markers may predict prognosis. Furthermore, it is as yet unclear if the use of multiple markers gives additional prognostic infomation. Cancer-associatedserum antigen (CASA) has been reported to have a high sensitivity9.Products of human chorionic gonadotrophin (hCG) have a prognostic
association with many tumours of the female reproductive tract'0.''. In this study we examined the prognostic value of PLAP, CA125, free P-hCG and CASA in women presenting with primary ovarian cancer.
METHODS
Correspondence: Mr T. Ind, Department of Obstetrics and Gynaecology, Watford Hospital, Vicarage Road, Watford WD 1 8HB, UK.
The study included 111 consecutive women with histologically confirmed epithelial ovarian cancer who had serum collected before initial surgery between 199'1 and 1994. Serum was stored at -20°C. Specimens were available for PLAP measurements in all cases and in 99, 99 and 73 cases for CA125, CASA and free P-hCG respectively. This was due to inadequate sample volumes for the latter three analytes. The characteristics of the 111 women are summarised in Table 1. Staging was performed according to the recommendations of the International Federation of Gynecology and Obstetrics (FIG0)l2 and the histological classificationused was that defined by the World Health Organization (WHO)I3.
1024
0 RCOG 1997 British Journal of Obstetrics and Gynaecology
1025
P R O G N O S T I C V A L U E O F CA12.5, C A S A , P L A P A N D F R E E P - H C G I N O V A R I A N C A N C E R
Table 1. Characteristics of the 111 women in the study. Values are given as n /total (%).
Age > 60 Cigarette smoker Histological type Serous Endometrioid Mucinous Clear cell Brenner Mixed Undifferentiated Positive tumour markers CAI25
CASA Free p-hCG PLAP
Unstaged
Stage 1
Stage 2
Stage 3
Stage 4
TOTAL
1 / 2 (50) 0 / 2 (0)
11 I 2 0 (55) 4 / 20 (20)
9 / 12 (75) 4 1 12 (33)
3 4 / 5 3 (64) 2 / 5 3 (4)
20 / 24 (83) 1/25 (4)
75/111 (68) I I / I l l (10)
(50) (10) (10) (7) (1) (4) (17)
0/2 012 012 012 012 112 112
(0) (0) (0) (0) (0) (50) (50)
4 / 20 4/20 5I20 3/20 1/20 2 1 10 1/20
(20) (20) (25) (15) (5) (10) (5)
5 / 12 1 112 2 / 12 4/12 0 1 12 0 1 12 0 1 12
(42) (8) (17) (33) (0) (0) (0)
35/53 5/53 4/53 0153 0153 1 f 53 8/53
(66) (9) (6) (0) (0) (2) (15)
12/24 1/24 1/24 I 124 0124 0124 9/24
(50) (4) (4) (4) (0) (0) (38)
561111 11/111 12/111 8/111 1/111 41111 19/111
Ill 011 o/ 1 1/ 2
(100) (0) (0) (50)
8/18 5 / 18 7/15 5 120
(44) (29) (47) (25)
7 / 1 1 (64) 3 / I I (27) 1 1 8 (13) 5 1 12 (42)
37 / 47 29 148 10/31 25 153
(79) (60) (32) (47)
17/22 16/23 8 1 18 14/24
(77) (70) (44) (58)
70199 (71) 53 199 (54) 26 173 (36) 50/111 (45)
CA125
Free B I C G
Serum concentrations of CA125 were measured in duplicate by immunoradiometric assay (Centocor Ltd, Malvern, Pennsylvania, USA). A cutoff level of 65 IU was used for comparison with other studies with similar control group^^^^^, and was based on the 92nd centile of values in 104 women with nonpregnancy-related, non-neoplastic or benign gynaecological disease.
Serum concentrations of free P-hCG were measured in duplicate using an in-house immunoradiometric assay which utilises the polyclonal antibody S752 (Polyclonal Antibodies Ltd, Dyfed, UK) for capture and radiolabelled monoclonal antibody 1/07 (Polyclonal Antibodies Ltd) for detection. The method has within and between assay variability of 1% to 5% and 7% to 9%, respectively. Cross-reactivity with free a-hCG, p core hCG, intact hCG, follicular stimulating hormone, luteinising hormone, and thyroid stimulating hormone were all < 0.1%. A cutoff level of 0.2 IU was used based on the 90th centile for a control group of 122 women with nonpregnancy related, nonmalignant or benign gynaecological disease.
PLAP Serum concentrations of PLAP were measured in duplicate by an in-house immunoradiometric assay, as previously describedI9.Concentrations were expressed as international units adjusted for smoking, ABO blood group and menopausal status20. A cutoff level of 7.4 adjusted units was used and is based on the 100th centile of 334 women with nonpregnancy related, non-neoplastic or benign gynaecological diseasei9. Univariant survival analysis for PLAP on 110 of the 111 patients has previously been reported5.
CASA Serum concentrations of tumour-associated mucin MUC-1 were determined in duplicate using the CASA assay which is a double determinant enzyme linked immunosorbent assay (ELISA) (Medical Innovations Ltd, Labrador, Australia). Cutoff values were based on smoking habits and age and were 3.7 IU, 5.5 IU,4.8 IU and 7-4 IU for nonsmokers and smokers aged younger than 60 years and those aged 60 years and older, respectively. These cutoff levels were based on the 95th centile of a control group of 4822 healthy women2' and were consistent with a control group of 106 women with nonpregnancy related, nonmalignant or benign gynaecologicaldisease. 0 RCOG 1997 Br J Obstet Gynaecol 104, 1024-1029
Statistics
The effect of each tumour marker on survival was analysed using the log rank x2 test stratified by disease stage22.The effects of each analyte on survival were also analysed separately with disease stage and in combination with stage, histological differentiation, age and smoking habits using the Cox proportional hazards model solved by the method of marginal l i k e l i h ~ o d ~ ~ .
RESULTS Survival curves for CA125, PLAP, CASA and free phCG are shown in Figs 1-4. Levels of all four markers were related to prognosis (stage corrected log rank x2 tests: CAI25 = 6.01, P = 0.01; PLAP = 28-65, P < 0.001; CASA x2 = 6-67, P = 0.01; free p-hCG = 13.44, P < 0.001). When fitted as single variables together with disease stage in Cox proportional hazard models, elevated levels
x2
x2
'x
1026
T. I N D E T A L .
80 Negative Y
5Q 40 -
I
n
P
a
“ C 20
2o
01 0
Positive
I
5
I
I
J 25
I
15 Survival (months) 10
20
Fig. 1. Survival plot of women with positive and negative serum CA 125 levels.
t
I
5
I
I
10 15 Survival (months)
I
20
I 25
Fig. 3. Survival plot of women with positive and negative serum CASA levels. 100
-8 80
m
Y
I
>
‘5 - 6o v)
0
2 40 5 m n 2
a
Positive
20
L
OO
5
10
15
20
25
a
I
I
I
I
I
I
5
10
15
20
25
Survival (months)
Fig. 2. Survival plot of women with positive and negative serum
PLAP levels.
Survival (months)
Fig. 4. Survival plot of women with positive and negative serum free p-hCG levels.
DISCUSSION of all four analytes were related to prognosis with the relative risk of mortality being greatest for PLAP (Table 2). When fitted as multiple variables in a Cox proportional hazards model, disease stage proved to be the single variable of greatest prognostic significance (Table 3). The combination of PLAP, CA125 and disease stage correlated with survival better than disease stage alone (Table 3). The addition of free P-hCG, CASA, age, tumour differentiation and smoking habits to the model did not demonstrate additional prognostic value (Table 3).
The prognostic value of CA125 and PLAP as individual measurements in women with primary epithelial ovarian cancer has been well do~umented*-~. Together with established factors such as stage, histological grade and age, these can help to predict survival. The present study confirms the prognostic value of serum CA125 and PLAP levels. These also had an association with survival when only the 1 3 women who had free P-hCG measurements were included in the analysis. Furthermore, these data demonstrate for the first time that serum concentrations of free P-hCG and CASA individually correlate with survival. However, 0 RCOG 1997 Br J Obstet Gynaecol 104, 1024-1029
PROGNOSTIC VALUE OF
CA125,
C A S A , PLAP A N D FREE P-HCG IN O V A R I A N C A N C E R
Table 2. Effects of prognostic factors when fitted as single variables together with disease stage in Cox proportional hazards models. Risk ratio: Exp(P)
Table 3. Effects of prognostic factors when fitted as multiple vanables in a Cox proportional hazards model. Risk ratio: Exp(P)
P
95% CI
~~
PLAP CA125 Free P-hCG CASA
4.10 2.43 2.3 1 1.90
2.27-7'37 1,234.77 1.314.10 1.05-3.42
< 0.0001 0,005 0.006 0.0274
the addition of free P-hCG and CASA measurements to the combination of disease stage, PLAP concentrations and CA 125 levels fails to provide additional prognostic information. In this study, the associations between the different markers were weak. Therefore, it could be argued that a larger study might demonstrate an additional association with survival for free P-hCG and CASA when combined with serum PLAP and CA125 concentrations. For the univariate analysis of free P-hCG, the power at a significance level of 0-05 is > 95%. However, the power of this study would be expected to decrease with the use of multivariate analysis. Furthermore, the power would be expected to decrease further if other established parameters such as performance status were included in the final model. In practical terms this large study failed to demonstrate any addition prognostic value for CASA and free P-hCG measurements. Therefore, there is no proven value for including them in a prognostic model in routine clinical practice. Elevated levels of numerous tumour-associated antigens are associated with prognosis in patients with epithelial ovarian carcinoma. In addition to those analytes described in the present study, comparable findings have been reported for serum concentrations of cross-linked carboxyterminal telopeptide of type 1 collagen24,interleukin 625,galactosyltransferase26,tumourassociated trypsin inhibitoS7,tetrane~tin~*,~~, sialyl Tn30, low levels of serum albumin3and expression of epidermal growth factoP'. Though serum concentrations of tumour-associated antigens are associated with disease this and other studies suggest that the prognostic significance is independent of tumour progression3y5. The effect may be due to increased protein transcription or tumour angiogenesis. Over-expression of oncogenes and transcription factors have been described in ovarian tumours with a poor p r o g n o ~ i s ~In ~ -addition, ~~, tumour angiogenesis may allow greater amounts of tumourassociated antigens to reach the circulation, and this itself is associated with poor survival in patients with ovarian carcinoma3'. Measurement of serum concentrations of ovarian tumour-associated antigens has been advocated for disease diagnosis and surveillance. Most information is available for CAI 2538which has the highest sensitivity 0 RCOG 1997 Br J Obstet Gynaecol 104, 1024-1029
1027
~~~
Stages 3 & 4 Elevated PLAP Elevated CA125 Elevated free P-hCG Age > 60 years Undifferentiated turnours Elevated CASA Cigarette smoking
~~
5.04 3.63 2-26 1.69 1.63 1-31 0.98 0.56
P
95% CI ~~~
1.96-1 2.92
1,93485 1.07-4.77 0.9 1-3. I4 0.85-3' 13 0.62-2.78 0.49-1 '96 0.09-3.52
~~
0~0001 < 0~0001 0.026 0.101
0.137 0.489 0.960 0.326
for ovarian cancer. In the present study sensitivity was 7 1%, consistent with other reports32.However, CAI 25 is not specific to ovarian neopla~ms~~. Therefore CAI25 does not contribute to clinical diagnosis. Similar problems apply to its use for ovarian cancer screening39. Probably the most valuable place for CA125 is in the follow up of women with proven disease'. Combined with surgical staging, serum CA125 concentrations may provide prognostic information. Furthermore, if debulking surgery for advanced ovarian cancer becomes obsolete, routine surgical staging will not be undertaken and the prognostic role of serological markers would assume greater importance. Although the sensitivity of CA125 is higher than that of other tumour markers, it may not be the best indicator of survival. The data presented here suggest that serum PLAP concentrations may be superior in terms of prognosis. However, comparisons of this type must be treated with caution. Risk estimates are biased by many other parameters used and the performance of the two analytes could not be distinguished statistically. The value of PLAP as a tumour marker is consistent with earlier studies by ourselves and others5,20,40-45. However, some authors report that PLAP is not an efficient marker of disease progression and s ~ r v i v a l ~ ~ * ~ ~ , ~ Some of these groups have used assays that measure the intrinsic catalytic activity of PLAP which is often defective when produced by t u m o ~ r s ~ ~The . ~ *assay . used in this study measured immunoreactivePLAP and is therefore not affected by this phenomenon40. Furthermore, there is an abundance of evidence that results should be corrected for blood group, menopausal status and smoking habit^^'^^^^^. The latter two variables should also be taken into account when using CASA as a tumour marker for epithelial ovarian ~ancet.2'.~~. In conclusion, this study confirms the prognostic value of preoperative serum CA125 and PLAP concentrations in women with primary epithelial ovarian carcinoma. We also report on the predictive value of CASA and free J3-hCG measurements. Concentrations of many serological tumour markers are associated with survival in ovarian carcinoma. However, which
1028 T. I N D
ET AL.
tumour-associated antigen is the best predictor; which combinations are best; and after the addition of how many tumour-associated antigens does further prognostic value cease have all yet to be defined. The present study suggests that serum PLAP concentrations are assocaited with survival better than other tumour-associated antigens. The study also shows that the addition of hCG and CASA to disease stage, CA125 and PLAP does not give additional prognostic information.
References 1 Rustin GJ. Nelstrop A. Stilwell J. Lambert HE. Savings obtained by
2
3
4 5
6 7 8 9
10
I1
12
13 14 15
16
CA-125 measurements during therapy for ovarian carcinoma. The North Thames Ovary Group. Eur J Cancer 1992; 28: 79-82. Nagele F, Petru E, Medi M, Kainz C, Graf AH, Sevelda P. Preoperative CA125: an independent prognostic factor in patients with stage I epithelial ovarian cancer. Obstet Gynecol 1995; 86: 259-264. Parker D, Bradley C, Bogle SM, Lay J, Masood M, Hancock AK, Naylor B, Price JJ. Serum albumin and CAI25 are powerful predictors of survival in epithelial ovarian cancer. Br J Obstet Gynaecol 1994; 101: 888-893. Davidson NG, Khanna S, Kinvan PH, Bircumshaw D. Prechemotherapy serum CAI25 level as a predictor of survival outcome in epithelial ovarian cancer. Clin Oncoll991; 3: 32-36. Ind TEJ, Iles RK, DeSouza K, Carter P, Lowe D, Shepherd JH, Hudson CN, Chard T. Serum placental-type alkaline phosphatase levels in patients with epithelial ovarian carcinoma. Int JOncol 1995; 6: 385-389. Parker D, Patel K, Aldred EJ, Hamden-mayor P, Naylor B. CA125 and survival in ovarian cancer: preliminary communication. J Roy SocMed 1991;81:22. Sevelda P, Schemper M, Spona J. CAI25 as an independantpredictor of survival in patients with epithelial ovarian cancer. Am J Obstet Gynecoll989; 97: 706712. Kenemans P, Yedema CA, Hilgers JHM et al. Clinical applications of monoclonal antibodies against ovarian cancer-associated antigens. Eur JOhstet Gynecol Reprod Bioll988; 29: 207-21 8. Devine PL, McCuckin MA, Ram LE, Ward BG, Pee D, Long S. Serum much antigens CASA and MSA in tumors of the breast, ovary, lung, pancreas, bladder, colon, and prostate. A blind trial with 420 patients. Cancer 1993; 72: 2007-2015. Carter PG, lles RK, Nevan P, Ind TEJ, Shepherd JH, Chard T. Measurements of urinary beta core fragment of human chorionic gonadotrophin in women with vulvovaginal malignancy and its prognostic significance.Er J Cancer 1995; 71: 350-353. Carter PG, Iles RK, Nevan P, Ind TEJ, Shepherd JH, Chard T. The prognostic significance of urinary p core fragment in premenopausal women with carcinoma of the cervix. Gynecol Oncol 1994; 55: 271-276. Pettersson F, Coppleson M, Creasman W, Ludwig H, Shepherd J. Annual Report on the Results of Treatment in GynaecologicalCancer, Edition 20. Stockholm: Fddtration lntemationale d’obstetrique et Gynkologie, 1988. Sarov SF, Scully RE, Sobin LH. International classification of tumours: 9. Histological typing of ovarian tumours. Geneva: World Health Organization, 1973. Niloff JM, Knapp RC, Schaetz E, Reynolds C, Bast RC. CA125 levels in obstetric and gynaecologic patients. Obstet Gynaecol 1984; 64: 703-707. Ricolleau G, Chatal J, Fumoleau P, Kremer M, Douillard J, Curtet C. Radioimmunoassay of the CAI25 antigen in ovarian carcinomas: advantages compared with CA19.9 and CEA. Tumour Biol 1984; 5: 151-159. Heinomen PK, Tontti K, Koivula T, Pystynen P. Tumor-associated antigen CAI 25 in patients with ovarian cancer. Br J Obstet Gynaecol 1985; 92: 528-531.
17 Einhorn N, Bast RC, Knapp RC, Tjemberg B, Zuraeski VR Preoperative evaluation of serum CAI25 levels in patients with primary epithelial ovarian cancer. Obstet Gynecoll986; 67: 41-16, 18 Fioretti P, Gadducci A, Ferdeghini M et al. Combined evaluation of some tumour associated antigens in the monitoring of integrated surgical and chemotherapeutic treatment of epithelial ovarian cancer. Eur JGynecolOncol1986; 7: 200-205. 19 Ind TEJ, Iles RK, Wathen NC et al. Low levels of amniotic fluid placental alkaline phosphatase in Down’s syndrome. Br J Obstet Gynaecoll993; 100: 847-849. 20 Ind TEJ, Iles RK, Carter PG et al. Serum placental-typealkaline phosphatase activity in women with squamous and glandular malignancies ofthe reproductive tract. JClin Patholl994; 47: 1035-1037. 21 McGuckin MA. Devine PL. Ramm LE. Ward BG. Factors effecting the measurement of tumor-associated h4UCl mucins in serum. TumourEiologv 1994; 15: 3 3 4 4 . 22 Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966; 50: 163-1 70. 23 Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure lime Data. New York: John Wiley, 1980. 24 Santala M, Risteli L, Puistola U, Risteli J, Kauppila A. Elevated serum ICTP concentrations reflect poor prognosis in patients with ovarian carcinoma. Ann Med 1995; 2 7 57-61. 25 Scambia G, Testa U, Benedetti P et al. Prognostic significance of interleukin 6 serum levels in patients with ovarian cancer. Br J Cancer 1995; 71: 354-356. 26 Sichel F, Salaun V, Bar E et al. Biological markers and ovarian carcinomas: galactosyltransferase, CA125, isoenzymes of amylase and alkaline phosphatase. Clin Chim Acta 1994; 227: 87-96. 27 Venesmaa P, Lehtovirta P, Stenman UH, Leminen A, Forss M, Ylikorkala 0. Tumour-associated trypsin inhibitor (TATI): comparison with CAI25 as a preoperative prognostic indicator in advanced ovarian cancer. Br JCancer 1994;70: 1188-1190. 28 Hogdall CK, Hogdall EV, Hording U, Clemmensen I, NorgaardPedersen B, Toftager-Larsen K. Pre-operative plasma tetranectin as a prognostic marker in ovarian cancer patients. &and J Clin Lab Invest 1993; 53: 741-746. 29 Hogdall CK, Christensen L, Clemmensen I. The prognostic value of tetranectin immunoreactivity and plasma tetranectin in patients with ovarian cancer. Cancer 1993; 72: 2415-2422. 30 Kobayashi H, Terao T, Kawashima Y. Serum sialyl Tn as an independent predictor of poor prognosis in patients with epithelial ovarian cancer. JCIin Oncoll992; 10: 95-101. 31 Kohler M, Kanz I, Wintzer HO, Wagner E, Bauknecht T. The expression of EGF receptors, EGF-like factors and c-myc in ovarian and cervical carcinoma and their potential clinical significance. AnticancerRes 1989; 9: 1537-1547. 32 Kenemans P, Bast RC, Yedema CA, Price MR, Hilgers J. CAI25 and polymorphic epithelial mucin as serum tumour markers. Cancer Rev 1988; 11: 119-144. 33 Levesque MA, Katsaros D, Yu H et al. Mutant p53 protein over expression is associated with poor outcome in patients with well or moderately differentiated ovarian carcinoma. Cancer 1995; 75: 1327-1 338. 34 Sheridan E, Silcocks P, Smith J, Hancock BW, Goyns MH. P53 mutation in a series of ovarian cancers from the UK, and its prognostic significance. Eur JCancer 1994; 30A: 1701-1704. 35 Singleton TP, Penone T, Oakley G et al. Activation of c-erbB-2 and prognosis in ovarian carcinoma. Comparison with histological type, grade and stage. Cancer 1994; 73: 1460-1466. 36 Henriksen R, Wilander E, Oberg K. Expression and prognostic significance of Bcl-2 in ovarian tumours. Er J Cancer 1995; 72: 1324-1329. 37 Hollingwrth HC, Kohn EC, Steinberg SM, Rothenberg ML, Merino MJ. Tumour angiogenesis in advanced stage ovarian carcinoma. Am J Patholl995; 147: 3341. 38 Bast RC, Feeney M, LWNS H, Nadler LM, Colvin RB,Knapp RC. Reactivity of a monoclonal antibody with human ovarian carcinoma. JClin Invest 1981;68: 1331-1337. 39 Prys-Davies A, Oram D. Screening for ovarian cancer. In: Studd J, editor. Progress in Obstetrics and Gynaecologv, Vol 9. London: Churchill Livingstone, 1991.
0 RCOG 1997 Br J Obstet Gynaecol 104, 1024-1029
P R O G N O S T I C V A L U E OF CA125, C A S A , P L A P A N D F R E E P - H C G IN O V A R I A N C A N C E R 1029 40 lles RK, Ind TEJ, Chard T. Production of placental alkaline phosphatase (PLAP) and PLAP-like material by epithelial germ cell and non-germ cell turnours in-vitro. Br J Cancer 1994;69:274-278. 41 Chatzaki E,Gallagher CJ, lles RK et al. Characterisation of the differential expression of marker antigens by normal and malignant endometrium. BrJCuncer 1994;69:1010-lO14. 42 DeBroe ME. Pollet DE. Multicenter evaluation of human placental alkaline phosphatase as a possible tumour-associated antigen in serum. CIin Chem 1988;34: 1995-1999. 43 Nozawa S, Udagawa Y,Ohkura H et al. Serum placental alkaline phosphatase (PLAP) in gynecologic malignancies-with special reference to the combination of PLAP and CA54/61 assay. CIin Chem Act0 1989;186:275-284. 44 Vergote I, Onsrud M, Nustad K. Placental alkaline phosphatase as a tumour marker in ovarian cancer. Ohstet Gynecol 1983;69:228-232. 45 Mano H,Furuhashi Y, Morikawa Y, Hattori SE, Goto S, Tomoda Y. Radioimmunoassay of placental alkaline phosphatase in ovarian cancer sera and tissues. Ohstet Gynecol 1986;68:759-764. 46 Haije WG, van Driel J, van der Burg ME. Catalytic and immunological activities of placental alkaline phosphatase in clinical studies. The value of PLAP in follow-up of ovarian cancer. Clin Chim Actu 1987; 165:165-175. 47 Doellpant GJ, Homesley HD. Placental-type alkaline phosphatase in ovarian cancer fluids and tissues. Obsrer Gynecol1984;63:324-329.
0 RCOG 1997 Br J Obstet Gynaecol 104, 1024-1029
48 Hoylaerts MF, Manes T, Millan JL. Molecular mechanisms of uncompetitive inhibition of human placental and gem-cell alkaline phosphatase. Biochem J 1992;286:23-30. 49 Tonik SE,Ortmeyer AE, Shendelman JE, Sussman HH. Elevation of serum placental alkaline phosphatase levels in cigarette smokers. Int J Cancer 1983;31:51-53. 50 Muensch H, Maslow WC, Azama F, Bertrand M, Dewhurst P, Hartrnan P. Placental alkaline phosphatase. Re-evaluation of the tumour marker after exclusion of smokers. Cancer 1986; 58: 1689-1 694. 5 I McLaughlin PJ,Twist AA, Evans CC, Johnson PM. Serum placental alkaline phosphatase in cigarette smokers. J Clin Puthol 1984;37: 826830. 52 Ind TEJ, lles RK, Chard T. Low levels of placental and placental-like alkaine phosphatase isoenzyines in women of blood groups A and AB.Ann Clin Chem 1994;31:89-90. 53 McGukin MA, Layton GT, Bailey MJ, Hunt T, Khoo SK, Ward BG. Evaluation of two new assays for tumour-associated antigens, CASA and OSA, found in the serum of patients with epithelial ovarian carcinoma-omparison with CA125. Gynecol Oncoll990;37: 165-17 I.
Received 3 December 1996 Accepted 14 April 1997
