Does serum CA-125 level prior to second-look laparotomy for invasive ovarian adenocarcinoma predict size of residual disease?

GYNECOLOGIC

ONCOLOGY

38, 373-376 (1990)

Does Serum CA-l 25 Level prior to Second-Look Laparotomy for Invasive Ovarian Adenocarcinoma Predict Size of Residual Disease? BRUCE PATSNER,

M.D.,“.’ JAMES W. ORR, JR., M.D .,* WILLIAM J. MANN, JR., M.D.,t

PEYTON T. TAYLOR, M.D.,+

EDWARD PARTRIDGE,M.D.,(( AND TARA ALLMEN, M.S.t *Division of Gynecologic Oncology, Watson Clinic, Lakeland, Florida 33804-5000; tDivision of Gynecologic Oncology, State University of New York at Stony Brook School of Medicine, Stony Brook, New York 11794-8091; *Division Gynecologic Oncology. University of Virginia School of Medicine, Charlottesville, Virginia 22903; and (ISouthern Gynecoiogic Oncology, P.C., Birmingham, Alabama 35205

Received January 23, 1990

known reference standard) prior to planned second-look

The records of 125 patients with nonmucinous invasive ovarian laparotomy virtually always indicates the presence of adenocarcinoma who underwent cytoreductive surgery, cisplatindisease, although a normal level is of limited value as at based combination chemotherapy, and second-look laparotomy were analyzed to correlate pre-second-look serum CA-125 levels least 50% of such patients will have residual disease as with the size of residual ovarian cancer. The majority of patients well [3,4]. Although virtually pathognomonic of the presence of with negative second-look laparotomy had normal serum CA-125 levels (46/50 or 92%). Of the 75 patients with positive second- residual adenocarcinoma when elevated prior to planned look, 56 (75%) had normal CA-125 levels preoperatively. Twentysecond-look surgery, it is unclear whether serum CAthree of twenty-four (96%) patients with residual diseaseless than 125 measurement alone or in combination with other or equal to 1 cm had normal CA-125 levels as did 20 of 28 (71%) techniques, such as CT scanning [51 or less invasive patients with disease 1.1-2.0 cm. Although elevated serum CA- surgical procedures such as laparoscopy, is an equivalent 125 levels were invariably associated with visible/gross disease substitute for a formal second-look surgical exploration and increasing size of residual disease tended to be associated in patients who have completed chemotherapy and have with increasing elevations of CA- 125, normal CA- 125 levels often no clinical evidence of disease. Concerns over possible occurred in the presenceof large-volume (>2 cm) disease(13/23, morbidity from full surgical exploration [6], controversy 57% of patients). The considerable overlap of serum CA-125 levels about the value of second cytoreductive surgery [7,8], for all sizes of residual disease precluded precise prediction of residual disease size based on serum CA-125 level alone. L 1990 limitations on the therapeutic armamentarium available Academic

Press. Inc.

CA-125 is a high-molecular-weight glycoprotein antigen expressed on the surface of cells derived from the coelomic epithelium, including over 80% of nonmucinous ovarian adenocarcinomas. This antigenic moiety is recognized by a monoclonal

antibody

designated

OC-125

which may be measured precisely in patients’ serum using established radioimmunometric techniques [ 11. The usefulness of serum CA-125 measurements in monitoring therapy of patients with invasive-ovarian adenocarcinoma has been firmly established in the past 5 years [2]. In particular, the presence of an elevated serum CA-125 (defined as greater than 35 U/ml measured against a Presented at the annual meeting of the Society of Gynecologic Oncologists, San Francisco, CA, February 4-7, 1990. ’ To whom reprint requests should be addressed.

to treat patients with large residual disease, and recognition that the recurrence rate following second-look surgery may be as high as 35-50% [9,101 are issues which complicate any attempt to find alternatives to full surgical exploration. There is, however, virtually universal agreement that the presence, and more importantly the size, of residual disease found at second-look surgery is of critical importance in predicting outcome [l 11. In particular, salvage regimens such as whole-abdomen radiation [ 121or intraperitoneal chemotherapy [ 131 are more successful when directed against microscopic residual disease and substantially less successful when residual disease is 2 cm or more in diameter. Although disease visualized at second-look laparoscopy is a valid assessment, the inherent inability of laparoscopy to evaluate the retroperitoneum and all parietal and visceral surfaces or to make

373 0090-8258190

$I 30

Copyright 0 1990by AcademicPress,Inc. All

rights

of reproduction

in any

form

reserved.

374

PATSNER ET AL.

precise estimates of size of disease throughout the peritoneal cavity limits its usefulness. As a result, any correlation of serum CA-125 levels with size of residual ovarian adenocarcinoma which is based wholly or in part on laparoscopic-derived data will be inherently inaccurate. The present prospective, multi-institutional study was undertaken to accurately and definitively assess the relationship between the preoperative level of serum CA125 and the size and location of residual disease.

to second-look surgery, and were not used to alter management decisions. Eleven patients had International Federation of Gynecology and Obstetrics (FIGO) stage I disease, eight stage II disease, ninety-five stage III disease, and eleven stage IV disease. Tumors were graded as either well, moderate, or poorly differentiated at each respective institution. No borderline malignancies were included in this series. RESULTS

METHODS AND MATERIALS Over a 36-month period from July 1986 to July 1989, 125 patients were consecutively entered from the institutions participating in this prospective study (School of Medicine, State University of New York at Stony Brook; Watson Clinic, Lakeland, FL; University of Virginia School of Medicine, Charlottesville; and Southern Gynecologic Oncology, P.C., Birmingham, AL). All patients had nonmucinous, invasive ovarian adenocarcinomas and elevated serum CA-125 levels at the time of initial cytoreductive surgery regardless of initial stage of disease. All patients had at least six courses of cisplatin-based combination chemotherapy and underwent full second-look laparotomy which included multiple washings, biopsies of all adhesions and peritoneal surfaces, diaphragm, residual omentum, or gastrococlic ligament, pelvic and paraaortic lymph node sampling, and resection of all suspicious lesions or tumor if technically feasible. Serum CA-125 levels were obtained on all patients immediately prior to each course of chemotherapy and second-look laparotomy. All patients had no clinical or radiographic evidence of disease prior to second-look surgery. No patient was excluded from second-look because of an elevated serum CA-125. All patients eligible for second-look underwent laparotomy, and all did so within 30 days of the last course of chemotherapy. Residual disease status was noted as the greatest diameter in centimeters of the largest tumor nodule found during laparotomy. Size was recorded in 0.5-cm increments to 1.0 cm, and in l-cm increments from l.l- to 5.0-cm intervals. Tumor found in blind peritoneal biopsies or cytologic washings was recorded as “microscopic,” and tumors larger than 5 cm were not quantified. Location of residual disease was also noted. All serum samples were evaluated using the same diagnostic laboratory kit for measurement of CA-125 levels. Investigators at each institution had access only to their own institutional data, and all clinical laboratory data were reviewed only by two authors (B.P. and T.A.). Levels were not known prior to each clinic visit or prior

One hundred twenty-five patients all underwent surgery, chemotherapy, and CA-125 surveillance as described. Fifty of the one hundred twenty-five patients (40%) had negative second-look laparotomies, and of these forty-six (92%) had normal levels and four (8%) had elevated levels. Seventy-five patients (60%) had a positive second-look laparotomy of which fifty-six patients (75%) had normal preoperative serum CA-125 levels and nineteen (25%) had elevated levels. Seventy-four of seventy-five patients (99%) with positive second-looks had intraperitoneal disease with or without retroperitoneal lymph node involvement. Only one patient (1%) had isolated nodal disease; this patient did have an elevated serum CA-125. The correlation of pre-second-look serum CA- 125 level with size of residual disease is listed in Table 1. Residual disease less than 1 cm in size is almost uniformly associated with a normal serum CA-125 level (23/24 patients, 96%). Seventy-one percent (20/28) of patients with disease 1.1 to 2.0 cm also had normal serum CA125 levels. Although tumor greater than 2.0 cm in diameter was generally associated with elevated serum CA-125 levels, 57% (13/23) of these patients had normal CA-125 levels. There was considerable overlap of the TABLE 1 Correlation of Serum CA-125 Level with Size of Residual Disease at Second-Look Laparotomy Size (cm) Negative Microscopic 0.1-0.5 0.6-I .O 1.1-2.0 2.1-3.0 3.1-4.0 4.1-5.0 25.1 Total

Total number of patients 50 7 12

Number with normal CA-125 46

Number with elevated CA-125 4 0

11

28 10 6

5 20 5 5

0 8 5

6

2

0 4

125

102

23

PRE-SECOND-LOOK

SERUM CA-125 AS PREDICTOR

TABLE 2 Respective Elevated Serum CA-125 Levels

for Differing Sizes

of Residual Disease

Size of disease

Total number of patients

Number with elevated CA-125

Percentage with elevated CA-125

Negative Microscopic 0.1-0.5 0.6- 1.O 1.1-2.0

50 7 12 5 28

4 0 I 0 8

8 0 8 0 29

2.1-3.0

IO

5

50

3. I-4.0 4.1-5.0 25. I

6 1 6

I 0 4

12 0 67

Values W/ml) 36, 45, 61, 125 67

40, 46, 48, 58, 65, 112, 600, 3800 54, 156, 217, 300, 4500 1000 49, 120, 150, 2000

range of serum CA-125 levels for all sizes of residual disease (Table 2). The sensitivity of CA-125 in the present study was only 0.27 whereas specificity of the test was 0.93. Positive predictive value was 0.83 and negative predictive value was only 0.49.

DISCUSSION

An elevated serum CA-125 level prior to planned second-look laparotomy may be almost equivalent to surgery in defining whether or not persistent adenocarcinoma will be found or become clinically evident in the next 4-6 months. The main value of second-look laparotomy is thus more precise determination of size and location of residual disease and the opportunity for secondary cytoreductive surgery. The latter, though, should realistically be performed only if residual disease is sufficiently large enough to warrant the anticipated morbidity from surgical exploration given the fact that all salvage regimens are significantly less effective with increasing size of disease. The current available literature suggests that pre-second-look serum CA-125 levels do correlate with size of residual disease. Were there a firm correlation between the degree of elevation of serum CA-125 and the size of residual disease and strong evidence that false-negative levels were accounted for only by small volume (i.e., microscopic or 2 cm) residual disease often is associated with a normal CA-125 (47% of patients in this series), and illustrates the considerable overlap of CA-125 values for all sizes of residual adenocarcinoma. This stands in direct contradiction to previous claims [3,4,15] that patients with “false-negative” CA-125 at the time of second-look laparotomy will invariably have small residual disease (less than 1 cm). Previously reported data [16] which suggested that small-volume disease is usually accompanied by normal CA-125 levels is confirmed in the present series. The discrepancy between our results and those of Niloff et a/. [3] and Berek et al. [4] is almost certainly due to the fact that all secondlook procedures in this investigation were laparotomies. The implications of the data in the present study are

376

PATSNER

clear. A given CA-125 level does not allow one to predict whether secondary debulking might be required, nor does it guarantee that salvage therapy will necessarily be directed at a preferred-size tumor volume. Serum CA125 only allows one to counsel patients about the likelihood of finding disease at the time a second-look operation might be performed, though probably no better than persistently elevated serum CA-125 levels after completion of three courses of chemotherapy. For those institutions or individuals who believe that salvage regimens are increasingly effective with increasingly smaller-volume disease, the serum CA-125 level alone will not allow one to decide whether secondary cytoreductive surgery will be necessary to achieve the requisite-size residual tumor volume or whether such already exists.

ET Al.. Another look at the second-assessment procedure for ovarian cpithelial carcinoma. Amer. J. Ohsrut. Gynecol. 157, 590-5Y6 (lY87). 7. Berck, J. S., Hacker. N. F.. I.agasse. 1.. D., Nieberg. R. K., and Elashoff. R. M. Survival of patients following secondary cytoreductive surgery in ovarian cancer. Oh.cfur. G~nerol. 61, 189-193 (1983). 8. I>ippman, S. M.. Alberts, D. S.. Slymen. D. J.. Weiner, S.. Aristizabal. S. A., Iuditch. A.. Davis, J. R., and Surwit, E. A. Second-look laparotomy in epithelial ovarian carcinoma prognostic factors associated with survival duration. Cancer 61, 2571-2577 t 198X). 9 Copeland. I.. J.. and Gcrshenson. D. M. Ovarian cancer recurrenccs in patients with no microscopic tumor at second-look laparotomy. Oh.rrcf. G.vneco/. 68, 873-874 (1986). IO Podratz. K. C.. Malkasian. (i. D.. Wicand. H. S.. Cha. S. S.. Lee, R. A.. Stanhope. C. R.. and Williams. T. J. Recurrent disease after negative second-look laparotomy in stages III and IV ovarian carcinoma. Gyftrc.o/. Orno/. 29, 274-282 f 1988). II

REFERENCES Bast. R. C., King. T. L., St. John, E.. Tenison. E.. Niloff. J.. Lazarus, H.. Berkowitz, R. S., Leavitt. T.. Griffiths. C. T.. Parker, L.. Zurawski, V. R.. and Knapp, R. C. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer, N. EngI. J. Med. 309, 883-887 (1983). Niloff. J. M. The role of the CA-125 assay in the management 01 ovarian cancer, Oncolo~.y 2, 67-72 (1988). Niloff. J. M.. Bast. R. C.. Jr., Schaetzl. E., and Knapp. R. C. Predictive value of CA-I?.( antigen levels in second-look procedures for ovarian cancer, Amer. J. Ohs/et. t;ynwol. 151, !#I-YX6 (1985).

I. Berek.

J. S.. Knapp, R. C., ney, C.. Niloff, J. M.. and correlated with second-look tients, Ohsrer. G.ynecol. 67,

Malkasian, G. D.. Lavin. P. T., WhitBast. R. C., Jr. CA-125 serum levels operations among ovarian cancer pa685-689 (1986).

Stehman, F. B., Calkins, A. R., Wass, J. I... Smirz. L. R.. Sutton. ci. I’., and Ehrlich. C. A comparison of findings at second-look laparotomy with preoperative computed tomography in patients with ovarian cancer, G.ynccol. Und. 29, 37-42 (1988). Gallup,

D. G., Muedo,

E.. Dudzinski.

M. R.. and Brown.

K. W.

.Miller. D. S.. Ballon. S. C.. Tcng. N. H.. Scifer. D. B.. and Soreiro. D. M. A critical reassessment of second-look laparotomy in epithelial ovarian cancer. Concur 57, 5NkS3.C (1986).

I2 Stephenson. J. A.. and Buchler. D. A. Abdominopclvic irradiation for pcrsistcnt advanced ovarian cancer. In!. J. Rudic~!. Oncd. Hid. f/?,v.S. 17, 385-388 (IY89). D. I... Emrich. L. J.. and 13 Pivcr. M. S.. Lclc. S. B.. Marchetti. Hartman. A. B. Surgically documented response to intraperitoneal cisplatin. cytarabine, and bleomycin after intravenous cisplatinhascd chemotherapy in advanced ovarian adenocarcinoma. J. C/in. OncrJ.

6. 1979-1984

(1988).

14. Rubin. S. C.. Hoskins. W. J.. Hakes. T. B., Markman. IM.. Reichmann. B. S.. Chapman. I).. and Icwis. J. L.. Jr. Serum CA-125 levels and surgical findings in patients undergoing secondary opcrations for cpithelial ovarian cancer. Anwr. J. Oh.stcat. Gynecd. 160, 667-771 (1989). IS. Komc. R. M.. Koh. H.. Fortune, D.. and Cauchi. M. CA-125 strum lcvcls and secondary laparotomy in epithclial ovarian tumors. ANSI. N.Z. J. Oh,src>t. Gynurcol. 27. 142-145 (IYX7). 16. Schilthuis. M. S.. Aaldcrs. J. G.. Bouma. H.. Kooi. H.. Fleuren, G. J.. Willemse, P. B.. and Dehruijn. H. W. A. Serum CA-12s levels in cpithelial ovarian cancer: Relation with findings at sccondlook operations and their role in the detection of tumor recurrence. Anwr.

J. Ohs/c~/. Gynecwl.

94, 202-207

(1987).