Risk of Malignancy Index in the preoperative evaluation of patients with adnexal masses

Available online at www.sciencedirect.com R

Gynecologic Oncology 90 (2003) 109 –112

www.elsevier.com/locate/ygyno

Risk of Malignancy Index in the preoperative evaluation of patients with adnexal masses Erik Soegaard Andersen, M.D., D.Msc.,a,* Aage Knudsen, M.D., D.MSc.,a Per Rix, M.D.,b and Birger Johansen, M.D.c a

Department of Obstetrics and Gynecology, Aalborg Hospital, Aalborg, Denmark Department of Obstetrics and Gynecology, Hjoerring Hospital, Hjoerring, Denmark c Department of Obstetrics and Gynecology, Frederikshavn Hospital, Frederikshavn, Denmark b

Received 25 September 2002

Abstract Objective. The aim of this study was to evaluate the use of a Risk of Malignancy Index (RMI) in primary evaluation of patients with adnexal masses. Methods. The RMI is based on menopausal status, ultrasonographic findings, and serum CA 125 level. A cutoff level of 200 was chosen as the threshold for referral for centralized primary surgery. This setup was evaluated by sensitivity, specificity, and positive predictive (PPV) and negative predictive (NPV) values with respect to the ability to distinguish malignant from benign pelvic masses. Results. The sensitivity was 70.6%, specificity 89.3%, PPV 66.1%, and NPV 91.1% for the total material. For the patients undergoing surgery the sensitivity was 70.6%, specificity 87.7%, PPV 66.1%, and NPV 89.8%. If stage I disease is considered “benign” disease, the sensitivity is 95.5%, specificity 87.9%, PPV 57.8%, and NPV 99.1%. Conclusions. RMI is a simple, easily applicable method in the primary evaluation of patients with adnexal masses. It is usable as a method for selective referral of relevant patients for centralized primary surgery. The method has significant limitations in borderline ovarian tumors, stage I invasive cancers, and nonepithelial tumors. Other methods should be evaluated to increase diagnostic accuracy in these cases. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Pelvic masses; Preoperative evaluation; Ovarian cancer

Introduction A woman presenting with an adenxal mass is a common clinical problem and a common reason for referral to hospital. The discrimination between benign and malignant tumors is an important step in the clinical handling of such cases. Up to 24% of ovarian tumors in premenopausal women are malignant and up to 60% are malignant in postmenopausal women [1–3]. The quality of primary cytoreductive surgery is one of the most important prognostic factors in ovarian cancer, and the extent of cytoreductive

* Corresponding author. Department of Obstetrics and Gynecology, Aalborg Hospital, Section North, DK-9000 Aalborg, Denmark. E-mail address: [email protected]; [email protected] (E.S. Andersen).

surgery is associated with the specific skills and experience of well-trained gynecologic oncologic surgeons [4,5]. The discrimination between benign and malignant adnexal masses is thus important in selective referral of relevant patients to specialized oncology centers. Used alone, the diagnostic accuracy of demographic, sonographic and biochemical variables is too poor for clinical use. The risk of malignancy index (RMI) was developed for referral of relevant patients to gynecologic oncologic centers [6]. The method has been validated prospectively and retrospectively in specialized and nonspecialized gynecologic departments [7–12]. The RMI is a simple scoring system based on menopausal status, ultrasonographic morphology, and serum CA 125 level. In our region, RMI was introduced as the method for primary evaluation and referral of patients with suspected

0090-8258/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0090-8258(03)00192-6

110

E.S. Andersen et al. / Gynecologic Oncology 90 (2003) 109 –112

malignant ovarian tumors from local hospitals to the specialized gynecologic oncologic center. The present study was performed to evaluate if RMI is able to select relevant patients for referral in our population.

Table 1 Malignant tumor types Epithelial ovarian cancer Stage I Stage II Stage III/IV Nonepithelial ovarian cancer Granulosa cell tumor Immature teratoma Sarcoma Metastatic tumors

Materials and methods The County of Northern Jutland, Denmark, comprises approximately 600.000 inhabitants. In 1999, centralization of primary surgery in ovarian cancer was agreed on by the regional gynecologic departments and politically approved. Patients with suspected malignant ovarian tumors were admitted to the Department of Obstetrics and Gynecology, Aalborg, for primary surgery. All women 30 years of age or older with a pelvic mass referred to the departments of gynecology in the County of Northern Jutland were evaluated with the RMI. The risk of malignancy index known as RMI 2 [7], with a cutoff level of 200, was decided to be the method distinguishing which patient should be referred for primary centralized surgery. Tingulstad et al. [7] created this model by a stepwise forward logistic regression, where serum CA 125 entered the model as a continous variable. The RMI was calculated as the product of the ultrasound score (U), the menopausal score (M), and the absolute value of serum CA 125. Multilocularity (more than bilocular), solid areas, bilaterality, ascites, and extraovarian tumors scored one point each. A total of 2 or more points gave U ⫽ 4; fewer than 2 points gave U ⫽ 1. Postmenopausal status was defined as more than 1 year of amenorrhea and scored M ⫽ 4; premenopausal status scored M ⫽ 1. Serum CA 125 was entered directly into the equation: RMI ⫽ U ⫻ M ⫻ CA125. Ultrasound was in all cases performed as a transvaginal examination; the abdominal approach was used when indicated. Preoperative serum CA 125 (IMX, Abbott Laboratories, USA) was determined, and menopausal status was registered. The RMI was calculated when the serum CA 125 result was present. If the RMI was greater than 200, the patient was referred to the Department of Obstetrics and Gynecology, Aalborg; if the RMI was less than 200 the patient underwent further treatment at the primary investigation site. In patients who underwent surgery, the surgical procedures, whether performed in the nonspecialized department or in the gynecologic oncologic center, were registered, and the results of examination of all histologic and cytologic specimens were registered. Patients who did not undergo surgery due to low RMI and small uni- or bilocular ovarian tumors and lack of symptoms were followed by repeat examination every 2 to 3 months. No patients were lost to follow-up. The RMI was evaluated for sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values with reference to the actual presence of a malignant or benign pelvic tumor. The histopathologic diagnosis was regarded

a

9I, 23Ba 13 47 2 1 1 6

I, invasive; B, borderline.

the true definite outcome. Patients who were observed after initial RMI scoring and who during follow-up had spontaneous regression of the ovarian tumor or who persistently had normal RMIs were considered to have a benign tumor.

Results From July 1999 to August 2001, 447 patients with a palpable or ultrasonographically demonstrated pelvic mass were consecutively evaluated with the RMI at the involved departments at the time of primary examination. In 402 cases (89.9%), patients underwent surgery. In 45 cases, the RMI was less than 200, a uni- or bilocular tumor was demonstrated, and the patients had no symptoms. In these cases, the patients were observed. In 23 cases (51.1%), the ovarian tumor regressed spontaneously. In 22 cases, repeat RMI every 2 to 3 months was persistently less than 200. None of these patients have been lost to follow-up and none have been diagnosed as having a malignant tumor during follow-up. In total, 102 patients (22.8%) had a malignant tumor, and the tumor types are listed in Table 1. A primary ovarian tumor was demonstrated in 96 cases (94.1%). In 92 cases (90.2%), a primary malignant epithelial tumor, including borderline ovarian tumors, was demonstrated. In 4 cases, a nonepithelial ovarian tumor was found, and in 6 cases, a secondary, metastatic tumor. All borderline ovarian tumors were stage I. Including borderline tumors, 34.8% were stage I, 14.1% were stage II, and 51.1% were stage III/IV. With respect to cases of invasive epithelial ovarian tumors, the stage distribution was 13.0% stage I, 18.8% stage II, and 68.1% stage III/IV. The secondary, metastatic tumors of the ovaries were carcinoid of the appendix; primary carcinomas of the pancreas, breast, and small intestine; a Krukenberg tumor; and one malignant lymphoma. The results of the preoperative evaluation by RMI are summarized in Table 2. Regarding the total material, including nonoperated patients, the sensitivity was 70.6% (72/102), the specificity 89.3% (308/345), the PPV 66.1% (72/109), and the NPV 91.1% (308/338). For the patients who underwent surgery, the sensitivity and PPV were exactly the same since all patients with a RMI ⬎ 200 under-

E.S. Andersen et al. / Gynecologic Oncology 90 (2003) 109 –112 Table 4 False-positive and false-negative cases

Table 2 Results of evaluation by RMI Benign Total material RMI ⬍ 200 RMI ⱖ 200 Surgically treated RMI ⬍ 200 RMI ⱖ 200

Malignant

308 37

30 72

263 37

30 72

went surgery. The specificity was 87.7% (263/300) and the NPV 89.8% (263/293). Since the purpose of initial evaluation of the patients by RMI was to ensure referral of relevant patients, namely, patients with advanced ovarian cancer, the results were analyzed considering stage I disease a “benign” disease. The results of this evaluation, for the total material, are summarized in Table 3. The sensitivity was then 95.5% (63/66), the specificity 87.9% (335/381), the PPV 57.8% (63/109), and the NPV 99.1% (335/338). Three patients (2.8%) with a RMI ⬎ 200 were not referred for primary surgical treatment at the subspecialized department. In one case, this was due to a misunderstanding of the rules for referral. In one case, the patient refused referral and was surgically treated by local gynecologists, and in one case, surgery was performed at the local hospital by a surgeon from the subspecialized center. In total, 37 cases were false positive (RMI ⬎ 200, benign tumor) and 30 cases were false negative (RMI ⬍ 200, malignant tumor). The false-positive and false-negative cases are listed in Table 4. Among false-positive cases, ovarian cystadenomas accounted for 45.9% and endometriosis for 27.0%. Infectious disease, tubo-ovarian abscess, and periappendicular abscess were found in 10.8%. The false-negative cases were primarily borderline ovarian tumors, 82.6% (19/23) having a RMI ⬍ 200; the 4 cases of nonepithelial ovarian tumors had a RMI ⬍ 200. In stage I invasive epithelial cancer 44.4% (4/9) had a RMI ⬍ 200. In more advanced invasive disease, stages II–IV, 5% (3/60) were false negative. All 4 cases of nonepithelial ovarian tumor had a RMI ⬍ 200.

Discussion The present study has demonstrated the usefulness of the RMI in prereferral evaluation of patients with demonstrated Table 3 Results of evaluation by RMI: total material

a

111

RMI

Benigna

Malignant

⬍200 ⱖ200

335 46

3 63

Stage I disease considered “benign”.

False-positive cases Cystadenoma Endometriosis Functional cyst Tuboovarian abscess Periappendicular abscess Hydrosalpinx Fibroma Dermoid cyst False-negative cases Borderline tumors Epithelial ovarian cancer stage I Epithelial ovarian cancer stage II Epithelial ovarian cancer stage III/IV Nonepithelial tumors

17 10 3 2 2 1 1 1 19 4 2 1 4

pelvic masses. The primary evaluation was performed in consecutively admitted patients in a similar setup despite the degree of specialization of the departments involved. The prevalence of malignancy in the present study (22.8%) and the stage distribution in cases of ovarian cancer indicate the unselected nature of the present material [2,6,9,13,14]. The primary purpose of introducing RMI into preoperative evaluation was the correct referral of patients with advanced ovarian cancer (greater than stage II) for primary surgery at the gynecologic oncologic center. Our results demonstrated that 95.5% were correctly identified prior to treatment. The correct identification of benign cases, as expressed by the specificity of the applied test, was approximately 90% This is in accordance with the results of other studies [7–12]. In this way, evaluation by RMI also prevented unnecessary referral of patients. Ideally, preoperative evaluation should be able to identify all cases of malignant tumor of the ovaries. Our results have demonstrated the limitations of RMI in identifying patients with stage I invasive disease and patients with ovarian tumors of borderline malignancy. Similar findings have been demonstrated in the previously cited reports. Due to the small number of stage I cancer patients, a more detailed analysis of substages in this group of patients was not relevant. In total, 5 of 9 (55.5%) stage I invasive cancers were correctly identified prior to surgery. Only 4 of 23 (17.4%) borderline ovarian tumors, all stage I, were correctly identified by RMI prior to surgery. Considering the biological nature and behavior of borderline ovarian tumors, in these patients, one can discuss the relevance of primary surgical treatment and staging at the subspecialized department. Borderline ovarian tumors are stage I tumors in more than 90% of cases and behave in a benign fashion in almost all cases. An argument in favour of aggressive surgical treatment by specialized oncologic surgeons is not reasonable in these cases. A more significant problem is the relatively low sensitivity in identifying patients with stage I invasive disease. Thorough surgical staging of these cases is an important part of the treatment strategy. The benefit of

112

E.S. Andersen et al. / Gynecologic Oncology 90 (2003) 109 –112

adjuvant treatment with chemotherapy in sufficiently staged patients with stage I disease has never been demonstrated [15]. Referral of all patients with stage I disease to surgeons skilled in performing optimal surgical staging would improve the selection of patients for further treatment. Further research is needed to evaluate methods able to identify these patients better than in the present and similar studies. A detailed evaluation of the present material is necessary, specifically to eventually improve the preoperative scoring. A very strong argument for the use of RMI in the primary evaluation of patients with pelvic masses is the simplicity of the method. The RMI can be used in daily clinical practice in nonspecialized gynecologic departments and by all gynecologists. We have not observed any problems with the method in the participating departments. Also, the high compliance, with almost 98% referred for centralized surgical treatment, demonstrates the applicability of RMI. It is also our impression that the use of RMI has resulted in better planning of the time of surgery, from referral to actual performance of surgery, and in planning of the surgical approach, laparoscopy versus laparotomy, transverse abdominal versus median abdominal incision. In conclusion, the present study has demonstrated the RMI to be a valuable, reliable, and applicable method in the primary evaluation of patients with pelvic masses and a usable method in referral of relevant patients for centralized surgical treatment. Other models of preoperative evaluation should be developed to improve the detection of nonepithelial ovarian cancers, borderline ovarian tumors, and earlystage invasive disease. Use of the methods in routine practice should be an important element of these methods.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

References [14] [1] Gilis CR, Hole DJ, Still RM, Davis J, Kaye SB. Medical audit, cancer registration, and survival in ovarian cancer. Lancet 1991;337:611–2. [2] Finkler NJ, Benacerraf B, Lavin PT, Wojcichhowski C, Knapp RC. Comparison of serum CA 125, clinical impression and ultrasound in

[15]

the preoperative evaluation of ovarian masses. Obstet Gynecol 1988; 72:659 – 64. Vasilev SA, Schlaerth JB, Campeau J, Morrow CP. Serum CA 125 levels in preoperative evaluation of pelvic masses. Obstet Gynecol 1988;71:751– 6. Eisenkop SM, Spirtos NM, Montag TW, Nalick RH, Wang H. The impact of subspeciality on the management of advanced ovarian cancer. Gynecol Oncol 1992;47:203–9. Kehoe S, Powell J, Wilson S, Woodman C. The influence of the operating surgeons specialisation on patient survival in ovarian carcinoma. Int J Cancer 1994;70:1014 –7. Iacobs I, Oram D, Fairbanks J, Turner J, Frost C, Grudzinskas JG. A Risk of Malignancy Index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol 1990;97:922–9. Tingulstad S, Hagen B, Skjeldestad FE, Onsrud M, Kiserud T, Halvorsen T, Nustad K. Evaluation of a risk of malignancy Index based on serum CA 125, ultrasound findings and menopausal status in the preoperative diagnosis of pelvic masses. Br J Obstet Gynaecol 1996; 103:826 –31. Tingulstad S, Hagen B, Skjeldestad FE, Halvorsen T, Nustad K, Onsrud M. The risk of Malignancy Index to evaluate potential ovarian cancers in local hospitals. Obstet Gynecol 1999;93:448 –52. Prys-Davies A, Iacobs I, Woolas R, Fish A, Oram D. The adnexal mass: benign or malignant? Evaluation of a Risk of Malignancy index. Br J Obstet Gynaecol 1993;100:927–31. Morgante G, la Marca A, Ditto A, De Leo V. Comparison of two malignancy indices based on serum CA 125, ultrasound score and menopausal status in the diagnosis of ovarian masses. Br J Obstet Gynaecol 1999;524 –7. Aslam N, Tailor A, Lawton F, Carr J, Savvas M, Jurkovic D. Prospective evaluation of three different models for the preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol 2000;107:1347– 53. Manjunath AP, Pratapkumar, Sujatha K, Vani R. Comparison of three risk of malignancy indices in evaluation of pelvic masses. Gynecol Oncol 2001;81:225–9. Malkasian GD, Knapp RC, Lavin PT, Zurawski VR Jr, Podratz KC. Preoperative evaluation of serum CA 125 levels in premenopausal and postmenopausal patients with pelvic masses: discrimination of benign from malignant disease. Am J Obstet Gynecol 1988;159: 341– 6. Petterson F. Annual report on the results of treatment in gynecological cancer. Int J Gynecol Cancer 1990;36(Suppl.):238 – 49. Favalli G, Odicino F, Torri V, Pecorelli S. Early stage ovarian cancer: the Italian contribution to clinical research. An update. Int J Gynecol Cancer 2001;11(Suppl. 1):12–9.