Original Article Oncol Res Treat 2015;38:276–281 DOI: 10.1159/000430858
Received: February 03, 2015 Accepted: April 21, 2015 Published online: May 19, 2015
The Risk of Ovarian Malignancy Algorithm (ROMA) as a Predictive Marker of Peritoneal Dissemination in Epithelial Ovarian Cancer Patients Yuji Ikeda a Kosei Hasegawa a Akira Kurosaki a Akiko Miyara a Tatsuya Hanaoka a Daisuke Shintani a Yuichi Imai a Tadaaki Nishikawa a Katsutoshi Oda b Keiichi Fujiwara a a Department
of Gynecologic Oncology, Saitama Medical University International Medical Center, Saitama, Japan; Department of Obstetrics and Gynecology, The University of Tokyo, Japan
Keywords Epithelial ovarian carcinoma · Peritoneal dissemination · Risk of ovarian malignancy algorithm Summary Background: This study aimed to determine the efficacy of the risk of ovarian malignancy algorithm (ROMA), calculated using the carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4) levels and the menopausal status, as a predictor of peritoneal dissemination in ovarian cancer. Methods: The CA125 and HE4 levels and the ROMA were compared between ovarian cancer patients (n = 122) with or without peritoneal dissemination. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated, and the results were compared with those of computed tomography (CT). Results: The CA125, HE4, and ROMA values differed significantly depending on the presence of peritoneal dissemination (p < 0.0001). The cut-off values were 181 U/ml for CA125, 161 pmol/ml for HE4, 44% for the ROMA (premenopausal), and 86% for the ROMA (postmenopausal). Among these markers, the ROMA (premenopausal) was the strongest predictor of peritoneal dissemination, with a specificity of 85.0% and a positive predictive value of 81.3%. In addition, the detection rates of small disseminations with less than 2 cm in diameter for the ROMA (93%) and HE4 (60%) were superior to that of CT (53%). Conclusions: The ROMA was a significant predictor of peritoneal dissemination and may be superior to CT for the detection of patients with small disseminations.
© 2015 S. Karger GmbH, Freiburg 2296–5270/15/0386–0276$39.50/0 Fax +49 761 4 52 07 14 [email protected] www.karger.com
Accessible online at: www.karger.com/ort
Introduction Ovarian cancer is the fifth leading cause of cancer-related death in women [1]. Diagnosis of this malignancy tends to be delayed because its symptoms are vague and often mimic other conditions. In fact, more than 60% of patients with ovarian cancer are diagnosed with advanced-stage disease and their prognosis is quite poor [1]. Complete surgical resection has been proven to be one of the most powerful survival determinants in the management of ovarian cancer patients [2]. Achievement of complete surgical resection is affected by several factors including the experience of the surgeon, the patient’s medical status, and the tumor locations. Identifying the existence of peritoneal dissemination before surgery is important in order to predict the possibilities of complete surgical resection [3]. Various therapeutic options such as neoadjuvant or intraperitoneal chemotherapy might be appropriate when complete resection is impossible [4]. Imaging by computed tomography (CT) is a powerful tool to detect peritoneal dissemination [5]. However, the cost of the equipment, radiation exposure, and limited sensitivity for small tumors are some disadvantages of CT [5, 6]. This is quite important because the wide range of small disseminations drastically decreases the complete resection rate [7]. Additionally, a history of allergic reaction to the contrast agent obviously decreases the diagnostic performance of CT [8]. Thus, alternative markers of peritoneal dissemination would be helpful in any situation. Carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4) are the commonly used biomarkers for epithelial ovarian cancer [9, 10]. Recently, Moore et al. [11, 12] published a series of papers in which they reported on the prediction of the presence of malignant
Kosei Hasegawa M.D., Ph.D. Department of Gynecologic Oncology Saitama Medical University International Medical Center 1397–1 Yamane Hidaka, Saitama 350–1298, Japan [email protected]
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
b
ovarian tumors by calculating the risk of ovarian malignancy algorithm (ROMA) score, which is based on the CA125 and HE4 levels and the menopausal status. This value showed notable diagnostic performance in distinguishing epithelial ovarian carcinoma (EOC) from benign ovarian tumors [13]. Although there is increasing evidence of the ROMA being a valuable marker for the prediction of malignancy, its usefulness as a predictive marker for peritoneal dissemination is limited [14, 15]. Therefore, the aim of this study is to determine appropriate ROMA, CA125, and HE4 cut-off values and to evaluate the efficacy of these markers as predictors of peritoneal dissemination.
Table 1. Patient characteristics Characteristic Age, years
Total Menopausal status Premenopausal Postmenopausal Histological subtype Serous Endometrioid Mucinous Clear Others FIGO stage I II III IV TNM pT1 pT2 pT3 pT3a pT3b pT3c pN pM
30–84
(median: 59)
n
%
122
100
35 87
29 71
51 20 5 31 15
42 16 4 25 12
40 17 50 15
33 14 41 12
40 20 62 1 14 47 9 15
33 16 51 2 23 76 7 12
Patients and Serum Samples Patients diagnosed with histologically confirmed EOC between December 2010 and March 2013, at the Saitama Medical University International Medical Center, were included in this study. Borderline tumors and cases after neoadjuvant chemotherapy were excluded. All patients provided informed consent for collection of samples and the use of their data, and this study was approved by the appropriate institutional ethics committee. Serum CA125 and HE4 levels were measured within a few days before the surgical procedure. Patients were considered to be postmenopausal if they were amenorrheic for more than 1 year, as determined by personal history taking. Tumors were pathologically confirmed and staged according to tumor/node/metastasis (TNM) criteria. CA125 and HE4 levels were measured using the ARCHITECT CA125II assay (Abbott Diagnostics, Abbott Park, IL, USA) and the HE4 EIA (Fujirebio Diagnostics AB, Gothenburg, Sweden), respectively, according to the manufacturers’ instructions. The ROMA predictive index (PI) was calculated using the following algorithms, as proposed by Moore et al. [11]: Premenopausal PI = –12 + 2.38 × LN[HE4] + 0.0626 × LN[CA125] Postmenopausal PI = –8.09 + 1.04 × LN[HE4] + 0.732 × LN[CA125] Predicted probability (ROMA score) = exp(PI)/[1 + exp(PI)] × 100 Imaging Diagnosis Diagnostic high-dose contrast-enhanced CT was performed and all images were reviewed by both an expert in imaging diagnosis and a gynecologic oncologist. Statistical Analysis The CA125, HE4, and ROMA values were compared between patients with peritoneal dissemination and those without peritoneal dissemination, using the Mann-Whitney U test (2-sided). The cut-off values of CA125, HE4, and the ROMA score were determined using receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) values. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for each marker and for imaging diagnosis. All results were considered significant at a level of 0.05. Statistical analyses were performed using JMP v11 (SAS, Cary, NC, USA).
Results Patient Characteristics The patient characteristics are shown in table 1. A total of 122 patients were included, with a median age of 59 years. 87 women (71%) were postmenopausal. With regard to tumor histology, most patients had serous adenocarcinoma and clear-cell adenocarcinoma (51 (42%) and 31 (25%) patients, respectively). Mixed and unclassified adenocarcinomas were classified as ‘others’. According
FIGO = International Federation of Gynecology and Obstetrics.
Table 2. Differences in serum marker values between patients with or without peritoneal disseminations
Methods
Dissemination
Mean
Median
SD
SE
Range
P value
CA125
pT1–pT2 pT3 pT1–pT2 pT3 pT1–pT2 pT3
606.7 9,173.3 256.4 789.3 46.7 90.8
110.1 943.6 83.7 570.2 33.1 97.6
376.8 75,790.4 376.8 647.9 36.8 17.3
48.6 6,647.3 48.6 82.3 4.8 2.2
4–21,634 4–865,591 33.8–1,649.9 47–2,625.3 3.0–99.8 7.4–99.9
< 0.0001
HE4 ROMA
< 0.0001 < 0.0001
ROMA as a Peritoneal Dissemination Marker in Ovarian Cancer
Oncol Res Treat 2015;38:276–281
277
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
SD = Standard deviation, SE = standard error.
Table 3. Predictive value of CT and each marker
Cut-off line
Sensitivity, %
Specificity, %
PPV, %
NPV, %
Contrast-enhanced CT
existence of pT3 suspected
77.4
88.8
87.3
79.1
CA125 HE4 ROMA (premenopausal) ROMA (postmenopausal)
181 U/ml ≥ 253 pmol/ml ≥ 44% ≥ 86%
93.5 79 86.7 89.4
61.7 80.0 85.0 62.5
71.6 80.3 81.3 73.7
90.2 78.7 89.5 83.3
Table 4. Preoperative evaluation of small (< 2 cm) disseminations in 15 cases
CT (identified before surgical procedure)
8 cases (53%)
HE4 (≥ 253 pmol/ml)
9 cases (60%)
ROMA (premenopausal) ≥ 44%/ (postmenopausal) 86%
14 cases (93%)
menopausal status. ROC analysis indicated that all markers had reasonable performance, with AUCs of 0.8026 for CA125, 0.8395 for HE4, 0.8733 for the ROMA (premenopausal), and 0.7968 for the ROMA (postmenopausal) (fig. 1). According to the Youden index, the appropriate cut-off values of CA125, HE4, ROMA (premenopausal), and ROMA (postmenopausal) to determine the presence of peritoneal dissemination were 197 U/ml, 161 pmol/ml, 86%, and 44%, respectively.
to surgical staging using the TNM criteria, 62 patients (51%) had peritoneal dissemination (pT3). Among these patients, 15 (25%) had disseminations of less than 2 cm (pT3a and pT3b). CA125, HE4, and the ROMA as Predictive Factors for Peritoneal Dissemination The CA125, HE4, and ROMA values in patients with (pT3) or without (pT1–pT2) peritoneal dissemination and pre of post-menopausal status are shown in table 2. The values of all markers differed significantly between patients in the pT1–pT2 groups and those in the pT3 group. The results of the analysis on the menopausal status showed that only the ROMA values were significantly different between the pre- and postmenopausal patients (p < 0.0001). Definition of the CA125, HE4, and ROMA Cut-Off Values We then verified the performance of CA125, HE4, and the ROMA in predicting peritoneal dissemination. Because the ROMA values were significantly different in pre- and postmenopausal patients, we used different cut-off ROMA values according to the
278
Oncol Res Treat 2015;38:276–281
Ikeda/Hasegawa/Kurosaki/Miyara/Hanaoka/ Shintani/Imai/Nishikawa/Oda/Fujiwara
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
Fig. 1. ROC curves of CA125, HE4, and the ROMA score (pre- and postmenopausal) for the prediction of peritoneal dissemination.
Prediction of Peritoneal Dissemination by CT, CA125, HE4, and the ROMA The sensitivity, specificity, PPV, and NPV of CT, CA125, HE4, and the ROMA for detecting the presence of peritoneal dissemination are shown in table 3. As expected, CT was able to predict the presence of peritoneal dissemination, with a specificity of 88.8% and a PPV of 87.3%. CA125, HE4, and the ROMA were analyzed using the identified cut-off values. The results using CA125 showed high sensitivity (93.5%), but sacrificing specificity (61.7%) and PPV (71.6%). The ROMA (premenopausal) showed the highest specificity (85.0%) and PPV (81.3%), which was comparable to the CT performances. On the other hand, the ROMA (postmenopausal) had poor specificity (62.5%) and a low PPV (73.7%), indicating the unsuitability of using ROMA to detect peritoneal dissemination in postmenopausal patients. The results of the HE4 analysis were acceptable, with a specificity of 80.0% and a PPV of 80.3%. As a result, the use of ROMA in patients with premenopausal status and HE4 regardless of the menopausal status was suitable for the prediction of peritoneal dissemination. Next, we focused on the patients with small peritoneal disseminations (< 2 cm in size; pT3a–pT3b) (table 4). Of the 15 pT3a– pT3b cases, only 8 cases (53%) were predicted to have peritoneal dissemination by preoperative CT. On the other hand, there were 9 (60%) and 14 (93%) cases above the cut-off lines of HE4 and ROMA, respectively, indicating the clinical significance of the ROMA as a predictive marker for small peritoneal disseminations.
Discussion In this study, the ROMA was assessed as a marker of peritoneal dissemination in patients with EOC. We established cut-off values of CA125, HE4, ROMA (premenopausal), and ROMA (postmenopausal) as 197 U/ml, 161 pmol/ml, 86%, and 44%, respectively. The results showed high specificity and a high PPV for predicting the presence of peritoneal dissemination. Although the vast majority of ovarian cancers occurs in postmenopausal women and is of advanced stage, a significant subset occurs in young women [16]. Therefore, the menopausal status should be carefully considered when analyzing serum markers because the ranges of their levels vary between pre- and postmenopausal patients. CA125 is influenced by several factors such as menopause, endometriosis, peritonitis, and pregnancy [17–19]. Therefore, CA125 might not be useful in the diagnostics of adnexal masses in premenopausal patients [20]. On the other hand, the manufacturer’s manual recommended to evaluate the results of HE4 regardless of the menopausal status [21]. The ROMA results are also known to be significantly different between pre- and postmenopausal women, and the normal cut-off value depends on the menopausal status [15, 22]. In our analysis, only the ROMA values were significantly different between pre- and postmenopausal patients with peritoneal dissemination. Notably, the specificity and PPV of the ROMA in premenopausal patients were superior to those in postmenopausal patients. This might be because the range of the ROMA in postmenopausal patients is narrow compared with those of other markers and patient characteristics. It is well known that CT is the standard imaging method for the preoperative evaluation of peritoneal dissemination [23]. However, CT is limited by low sensitivity to detect small size tumors; for tumors of less than 1 cm in diameter, the sensitivity is only 25–50% [5]. In our study, preoperative prediction of peritoneal disseminations of less than 2 cm using CT was inferior to the ROMA, demonstrating that the ROMA is a useful biomarker to predict the presence of small peritoneal disseminations.
This study had some limitations. In particular, this was a retrospective study, and a small number of patients were included. However, serum marker tests have the advantage of being inexpensive, noninvasive, and routinely performed in most facilities.
Conclusions In conclusion, we established the appropriate cut-off values for the ROMA score, CA125, and HE4. Our study suggests that the utility of the ROMA as a marker of disease progression should be further explored in larger studies, especially in premenopausal patients, as it may be superior to CT for the detection of disseminations of less than 2 cm.
Acknowledgements We thank Yuko Ijima and Akiko Iwasa for their excellent technical assistance. We also thank Kenji Abe and Dr. Tetsuo Sekino, EIDIA Co., Ltd., for their support. We appreciate the kind help of Dr. Houda Alachkar for English editing.
Authors’ Contributions Y.Ik. carried out the data collection and drafted the manuscript. K.H. participated in the design of the study. A.K. carried out the statistical analysis. T.H., D.S., and Y.Im. collected the clinical samples. Y.Im. and T.N. carried out the data collection. K.O. participated in the design of the study and reading the proofs of the manuscript. K.F. participated in the design of the study.
Disclosure Statement All the authors declare no financial competing interests.
References
280
4 Fujiwara K, Nagao S, Aotani E, Hasegawa K: Principle and evolving role of intraperitoneal chemotherapy in ovarian cancer. Expert Opin Pharmacother 2013; 14: 1797–1806. 5 Coakley FV, Choi PH, Gougoutas CA, Pothuri B, Venkatraman E, Chi D, Bergman A, Hricak H: Peritoneal metastases: detection with spiral CT in patients with ovarian cancer. Radiology 2002; 223: 495–499. 6 Patel CM, Sahdev A, Reznek RH: CT, MRI and PET imaging in peritoneal malignancy. Cancer Imaging 2011; 11: 123–139. 7 Sehouli J, Senyuva F, Fotopoulou C, Neumann U, Denkert C, Werner L, Gulten OO: Intra-abdominal tumor dissemination pattern and surgical outcome in 214 patients with primary ovarian cancer. J Surg Oncol 2009; 99: 424–427. 8 Davis KR, New PF, Solis OJ, Roberson GH: A review of the findings on computed cranial tomography following intravenous contrast media. Rev Interam Radiol 1977; 2: 15–18.
Oncol Res Treat 2015;38:276–281
9 Partridge E, Kreimer AR, Greenlee RT, Williams C, Xu JL, Church TR, Kessel B, Johnson CC, Weissfeld JL, Isaacs C, Andriole GL, Ogden S, Ragard LR, Buys SS, Team PP: Results from four rounds of ovarian cancer screening in a randomized trial. Obstet Gynecol 2009; 113: 775–782. 10 Hellstrom I, Raycraft J, Hayden-Ledbetter M, Ledbetter JA, Schummer M, McIntosh M, Drescher C, Urban N, Hellstrom KE: The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer Res 2003; 63: 3695–3700. 11 Moore RG, McMeekin DS, Brown AK, DiSilvestro P, Miller MC, Allard WJ, Gajewski W, Kurman R, Bast RC Jr, Skates SJ: A novel multiple marker bioassay utilizing HE4 and CA125 for the prediction of ovarian cancer in patients with a pelvic mass. Gynecol Oncol 2009; 112: 40–46.
Ikeda/Hasegawa/Kurosaki/Miyara/Hanaoka/ Shintani/Imai/Nishikawa/Oda/Fujiwara
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
1 Siegel R, Naishadham D, Jemal A: Cancer statistics, 2013. CA Cancer J Clin 2013; 63: 11–30. 2 Wimberger P, Lehmann N, Kimmig R, Burges A, Meier W, Du Bois A; Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study Group: Prognostic factors for complete debulking in advanced ovarian cancer and its impact on survival. An exploratory analysis of a prospectively randomized phase III study of the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study Group (AGOOVAR). Gynecol Oncol 2007; 106: 69–74. 3 Fotopoulou C, Richter R, Braicu EI, Schmidt SC, Lichtenegger W, Sehouli J: Can complete tumor resection be predicted in advanced primary epithelial ovarian cancer? A systematic evaluation of 360 consecutive patients. Eur J Surg Oncol 2010; 36: 1202–1210.
ROMA as a Peritoneal Dissemination Marker in Ovarian Cancer
16 Gershenson DM: Treatment of ovarian cancer in young women. Clin Obstet Gynecol 2012; 55: 65–74. 17 Fedele L, Vercellini P, Arcaini L, da Dalt MG, Candiani GB: CA 125 in serum, peritoneal fluid, active lesions, and endometrium of patients with endometriosis. Am J Obstet Gynecol 1988; 158: 166–170. 18 Panorchan K, Davenport A: Diagnostic and prognostic role of peritoneal CA 125 in peritoneal dialysis patients presenting with acute peritonitis. BMC Nephrol 2014; 15: 149. 19 Check JH, Vetter BH: A challenge to the concept of tubal reflux to explain the rise and fall of CA125 in serum during the first trimester. Hum Reprod 1995;10: 674–676. 20 Holcomb K, Vucetic Z, Miller MC, Knapp RC: Human epididymis protein 4 offers superior specificity in the differentiation of benign and malignant adnexal masses in premenopausal women. Am J Obstet Gynecol 2011; 205: 358.e1–e6.
21 Anastasi E, Granato T, Marchei GG, Viggiani V, Colaprisca B, Comploj S, Reale MG, Frati L, Midulla C: Ovarian tumor marker HE4 is differently expressed during the phases of the menstrual cycle in healthy young women. Tumour Biol 2010; 31: 411–415. 22 Montagnana M, Danese E, Ruzzenente O, Bresciani V, Nuzzo T, Gelati M, Salvagno GL, Franchi M, Lippi G, Guidi GC: The ROMA (risk of ovarian malignancy algorithm) for estimating the risk of epithelial ovarian cancer in women presenting with pelvic mass: Is it really useful? Clin Chem Lab Med 2011; 49: 521–525. 23 Manegold-Brauer G, Bellin AK, Tercanli S, Lapaire O, Heinzelmann-Schwarz V: The special role of ultrasound for screening, staging and surveillance of malignant ovarian tumors: distinction from other methods of diagnostic imaging. Arch Gynecol Obstet 2014; 289: 491–498.
Oncol Res Treat 2015;38:276–281
281
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
12 Moore RG, Jabre-Raughley M, Brown AK, Robison KM, Miller MC, Allard WJ, Kurman RJ, Bast RC, Skates SJ: Comparison of a novel multiple marker assay vs the risk of malignancy index for the prediction of epithelial ovarian cancer in patients with a pelvic mass. Am J Obstet Gynecol 2010; 203: 228.e1–e6. 13 Li F, Tie R, Chang K, Wang F, Deng S, Lu W, Yu L, Chen M: Does risk for ovarian malignancy algorithm excel human epididymis protein 4 and CA125 in predicting epithelial ovarian cancer: a meta-analysis. BMC Cancer 2012; 12: 258. 14 Karlsen MA, Sandhu N, Hogdall C, Christensen IJ, Nedergaard L, Lundvall L, Engelholm SA, Pedersen AT, Hartwell D, Lydolph M, Laursen IA, Hogdall EV: Evaluation of HE4, CA125, risk of ovarian malignancy algorithm (ROMA) and risk of malignancy index (RMI) as diagnostic tools of epithelial ovarian cancer in patients with a pelvic mass. Gynecol Oncol 2012; 127: 379–383. 15 Sandri MT, Bottari F, Franchi D, Boveri S, Candiani M, Ronzoni S, Peiretti M, Radice D, Passerini R, Sideri M: Comparison of HE4, CA125 and ROMA algorithm in women with a pelvic mass: correlation with pathological outcome. Gynecol Oncol 2013; 128: 233–238.
Received: February 03, 2015 Accepted: April 21, 2015 Published online: May 19, 2015
The Risk of Ovarian Malignancy Algorithm (ROMA) as a Predictive Marker of Peritoneal Dissemination in Epithelial Ovarian Cancer Patients Yuji Ikeda a Kosei Hasegawa a Akira Kurosaki a Akiko Miyara a Tatsuya Hanaoka a Daisuke Shintani a Yuichi Imai a Tadaaki Nishikawa a Katsutoshi Oda b Keiichi Fujiwara a a Department
of Gynecologic Oncology, Saitama Medical University International Medical Center, Saitama, Japan; Department of Obstetrics and Gynecology, The University of Tokyo, Japan
Keywords Epithelial ovarian carcinoma · Peritoneal dissemination · Risk of ovarian malignancy algorithm Summary Background: This study aimed to determine the efficacy of the risk of ovarian malignancy algorithm (ROMA), calculated using the carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4) levels and the menopausal status, as a predictor of peritoneal dissemination in ovarian cancer. Methods: The CA125 and HE4 levels and the ROMA were compared between ovarian cancer patients (n = 122) with or without peritoneal dissemination. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated, and the results were compared with those of computed tomography (CT). Results: The CA125, HE4, and ROMA values differed significantly depending on the presence of peritoneal dissemination (p < 0.0001). The cut-off values were 181 U/ml for CA125, 161 pmol/ml for HE4, 44% for the ROMA (premenopausal), and 86% for the ROMA (postmenopausal). Among these markers, the ROMA (premenopausal) was the strongest predictor of peritoneal dissemination, with a specificity of 85.0% and a positive predictive value of 81.3%. In addition, the detection rates of small disseminations with less than 2 cm in diameter for the ROMA (93%) and HE4 (60%) were superior to that of CT (53%). Conclusions: The ROMA was a significant predictor of peritoneal dissemination and may be superior to CT for the detection of patients with small disseminations.
© 2015 S. Karger GmbH, Freiburg 2296–5270/15/0386–0276$39.50/0 Fax +49 761 4 52 07 14 [email protected] www.karger.com
Accessible online at: www.karger.com/ort
Introduction Ovarian cancer is the fifth leading cause of cancer-related death in women [1]. Diagnosis of this malignancy tends to be delayed because its symptoms are vague and often mimic other conditions. In fact, more than 60% of patients with ovarian cancer are diagnosed with advanced-stage disease and their prognosis is quite poor [1]. Complete surgical resection has been proven to be one of the most powerful survival determinants in the management of ovarian cancer patients [2]. Achievement of complete surgical resection is affected by several factors including the experience of the surgeon, the patient’s medical status, and the tumor locations. Identifying the existence of peritoneal dissemination before surgery is important in order to predict the possibilities of complete surgical resection [3]. Various therapeutic options such as neoadjuvant or intraperitoneal chemotherapy might be appropriate when complete resection is impossible [4]. Imaging by computed tomography (CT) is a powerful tool to detect peritoneal dissemination [5]. However, the cost of the equipment, radiation exposure, and limited sensitivity for small tumors are some disadvantages of CT [5, 6]. This is quite important because the wide range of small disseminations drastically decreases the complete resection rate [7]. Additionally, a history of allergic reaction to the contrast agent obviously decreases the diagnostic performance of CT [8]. Thus, alternative markers of peritoneal dissemination would be helpful in any situation. Carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4) are the commonly used biomarkers for epithelial ovarian cancer [9, 10]. Recently, Moore et al. [11, 12] published a series of papers in which they reported on the prediction of the presence of malignant
Kosei Hasegawa M.D., Ph.D. Department of Gynecologic Oncology Saitama Medical University International Medical Center 1397–1 Yamane Hidaka, Saitama 350–1298, Japan [email protected]
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
b
ovarian tumors by calculating the risk of ovarian malignancy algorithm (ROMA) score, which is based on the CA125 and HE4 levels and the menopausal status. This value showed notable diagnostic performance in distinguishing epithelial ovarian carcinoma (EOC) from benign ovarian tumors [13]. Although there is increasing evidence of the ROMA being a valuable marker for the prediction of malignancy, its usefulness as a predictive marker for peritoneal dissemination is limited [14, 15]. Therefore, the aim of this study is to determine appropriate ROMA, CA125, and HE4 cut-off values and to evaluate the efficacy of these markers as predictors of peritoneal dissemination.
Table 1. Patient characteristics Characteristic Age, years
Total Menopausal status Premenopausal Postmenopausal Histological subtype Serous Endometrioid Mucinous Clear Others FIGO stage I II III IV TNM pT1 pT2 pT3 pT3a pT3b pT3c pN pM
30–84
(median: 59)
n
%
122
100
35 87
29 71
51 20 5 31 15
42 16 4 25 12
40 17 50 15
33 14 41 12
40 20 62 1 14 47 9 15
33 16 51 2 23 76 7 12
Patients and Serum Samples Patients diagnosed with histologically confirmed EOC between December 2010 and March 2013, at the Saitama Medical University International Medical Center, were included in this study. Borderline tumors and cases after neoadjuvant chemotherapy were excluded. All patients provided informed consent for collection of samples and the use of their data, and this study was approved by the appropriate institutional ethics committee. Serum CA125 and HE4 levels were measured within a few days before the surgical procedure. Patients were considered to be postmenopausal if they were amenorrheic for more than 1 year, as determined by personal history taking. Tumors were pathologically confirmed and staged according to tumor/node/metastasis (TNM) criteria. CA125 and HE4 levels were measured using the ARCHITECT CA125II assay (Abbott Diagnostics, Abbott Park, IL, USA) and the HE4 EIA (Fujirebio Diagnostics AB, Gothenburg, Sweden), respectively, according to the manufacturers’ instructions. The ROMA predictive index (PI) was calculated using the following algorithms, as proposed by Moore et al. [11]: Premenopausal PI = –12 + 2.38 × LN[HE4] + 0.0626 × LN[CA125] Postmenopausal PI = –8.09 + 1.04 × LN[HE4] + 0.732 × LN[CA125] Predicted probability (ROMA score) = exp(PI)/[1 + exp(PI)] × 100 Imaging Diagnosis Diagnostic high-dose contrast-enhanced CT was performed and all images were reviewed by both an expert in imaging diagnosis and a gynecologic oncologist. Statistical Analysis The CA125, HE4, and ROMA values were compared between patients with peritoneal dissemination and those without peritoneal dissemination, using the Mann-Whitney U test (2-sided). The cut-off values of CA125, HE4, and the ROMA score were determined using receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) values. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for each marker and for imaging diagnosis. All results were considered significant at a level of 0.05. Statistical analyses were performed using JMP v11 (SAS, Cary, NC, USA).
Results Patient Characteristics The patient characteristics are shown in table 1. A total of 122 patients were included, with a median age of 59 years. 87 women (71%) were postmenopausal. With regard to tumor histology, most patients had serous adenocarcinoma and clear-cell adenocarcinoma (51 (42%) and 31 (25%) patients, respectively). Mixed and unclassified adenocarcinomas were classified as ‘others’. According
FIGO = International Federation of Gynecology and Obstetrics.
Table 2. Differences in serum marker values between patients with or without peritoneal disseminations
Methods
Dissemination
Mean
Median
SD
SE
Range
P value
CA125
pT1–pT2 pT3 pT1–pT2 pT3 pT1–pT2 pT3
606.7 9,173.3 256.4 789.3 46.7 90.8
110.1 943.6 83.7 570.2 33.1 97.6
376.8 75,790.4 376.8 647.9 36.8 17.3
48.6 6,647.3 48.6 82.3 4.8 2.2
4–21,634 4–865,591 33.8–1,649.9 47–2,625.3 3.0–99.8 7.4–99.9
< 0.0001
HE4 ROMA
< 0.0001 < 0.0001
ROMA as a Peritoneal Dissemination Marker in Ovarian Cancer
Oncol Res Treat 2015;38:276–281
277
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
SD = Standard deviation, SE = standard error.
Table 3. Predictive value of CT and each marker
Cut-off line
Sensitivity, %
Specificity, %
PPV, %
NPV, %
Contrast-enhanced CT
existence of pT3 suspected
77.4
88.8
87.3
79.1
CA125 HE4 ROMA (premenopausal) ROMA (postmenopausal)
181 U/ml ≥ 253 pmol/ml ≥ 44% ≥ 86%
93.5 79 86.7 89.4
61.7 80.0 85.0 62.5
71.6 80.3 81.3 73.7
90.2 78.7 89.5 83.3
Table 4. Preoperative evaluation of small (< 2 cm) disseminations in 15 cases
CT (identified before surgical procedure)
8 cases (53%)
HE4 (≥ 253 pmol/ml)
9 cases (60%)
ROMA (premenopausal) ≥ 44%/ (postmenopausal) 86%
14 cases (93%)
menopausal status. ROC analysis indicated that all markers had reasonable performance, with AUCs of 0.8026 for CA125, 0.8395 for HE4, 0.8733 for the ROMA (premenopausal), and 0.7968 for the ROMA (postmenopausal) (fig. 1). According to the Youden index, the appropriate cut-off values of CA125, HE4, ROMA (premenopausal), and ROMA (postmenopausal) to determine the presence of peritoneal dissemination were 197 U/ml, 161 pmol/ml, 86%, and 44%, respectively.
to surgical staging using the TNM criteria, 62 patients (51%) had peritoneal dissemination (pT3). Among these patients, 15 (25%) had disseminations of less than 2 cm (pT3a and pT3b). CA125, HE4, and the ROMA as Predictive Factors for Peritoneal Dissemination The CA125, HE4, and ROMA values in patients with (pT3) or without (pT1–pT2) peritoneal dissemination and pre of post-menopausal status are shown in table 2. The values of all markers differed significantly between patients in the pT1–pT2 groups and those in the pT3 group. The results of the analysis on the menopausal status showed that only the ROMA values were significantly different between the pre- and postmenopausal patients (p < 0.0001). Definition of the CA125, HE4, and ROMA Cut-Off Values We then verified the performance of CA125, HE4, and the ROMA in predicting peritoneal dissemination. Because the ROMA values were significantly different in pre- and postmenopausal patients, we used different cut-off ROMA values according to the
278
Oncol Res Treat 2015;38:276–281
Ikeda/Hasegawa/Kurosaki/Miyara/Hanaoka/ Shintani/Imai/Nishikawa/Oda/Fujiwara
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
Fig. 1. ROC curves of CA125, HE4, and the ROMA score (pre- and postmenopausal) for the prediction of peritoneal dissemination.
Prediction of Peritoneal Dissemination by CT, CA125, HE4, and the ROMA The sensitivity, specificity, PPV, and NPV of CT, CA125, HE4, and the ROMA for detecting the presence of peritoneal dissemination are shown in table 3. As expected, CT was able to predict the presence of peritoneal dissemination, with a specificity of 88.8% and a PPV of 87.3%. CA125, HE4, and the ROMA were analyzed using the identified cut-off values. The results using CA125 showed high sensitivity (93.5%), but sacrificing specificity (61.7%) and PPV (71.6%). The ROMA (premenopausal) showed the highest specificity (85.0%) and PPV (81.3%), which was comparable to the CT performances. On the other hand, the ROMA (postmenopausal) had poor specificity (62.5%) and a low PPV (73.7%), indicating the unsuitability of using ROMA to detect peritoneal dissemination in postmenopausal patients. The results of the HE4 analysis were acceptable, with a specificity of 80.0% and a PPV of 80.3%. As a result, the use of ROMA in patients with premenopausal status and HE4 regardless of the menopausal status was suitable for the prediction of peritoneal dissemination. Next, we focused on the patients with small peritoneal disseminations (< 2 cm in size; pT3a–pT3b) (table 4). Of the 15 pT3a– pT3b cases, only 8 cases (53%) were predicted to have peritoneal dissemination by preoperative CT. On the other hand, there were 9 (60%) and 14 (93%) cases above the cut-off lines of HE4 and ROMA, respectively, indicating the clinical significance of the ROMA as a predictive marker for small peritoneal disseminations.
Discussion In this study, the ROMA was assessed as a marker of peritoneal dissemination in patients with EOC. We established cut-off values of CA125, HE4, ROMA (premenopausal), and ROMA (postmenopausal) as 197 U/ml, 161 pmol/ml, 86%, and 44%, respectively. The results showed high specificity and a high PPV for predicting the presence of peritoneal dissemination. Although the vast majority of ovarian cancers occurs in postmenopausal women and is of advanced stage, a significant subset occurs in young women [16]. Therefore, the menopausal status should be carefully considered when analyzing serum markers because the ranges of their levels vary between pre- and postmenopausal patients. CA125 is influenced by several factors such as menopause, endometriosis, peritonitis, and pregnancy [17–19]. Therefore, CA125 might not be useful in the diagnostics of adnexal masses in premenopausal patients [20]. On the other hand, the manufacturer’s manual recommended to evaluate the results of HE4 regardless of the menopausal status [21]. The ROMA results are also known to be significantly different between pre- and postmenopausal women, and the normal cut-off value depends on the menopausal status [15, 22]. In our analysis, only the ROMA values were significantly different between pre- and postmenopausal patients with peritoneal dissemination. Notably, the specificity and PPV of the ROMA in premenopausal patients were superior to those in postmenopausal patients. This might be because the range of the ROMA in postmenopausal patients is narrow compared with those of other markers and patient characteristics. It is well known that CT is the standard imaging method for the preoperative evaluation of peritoneal dissemination [23]. However, CT is limited by low sensitivity to detect small size tumors; for tumors of less than 1 cm in diameter, the sensitivity is only 25–50% [5]. In our study, preoperative prediction of peritoneal disseminations of less than 2 cm using CT was inferior to the ROMA, demonstrating that the ROMA is a useful biomarker to predict the presence of small peritoneal disseminations.
This study had some limitations. In particular, this was a retrospective study, and a small number of patients were included. However, serum marker tests have the advantage of being inexpensive, noninvasive, and routinely performed in most facilities.
Conclusions In conclusion, we established the appropriate cut-off values for the ROMA score, CA125, and HE4. Our study suggests that the utility of the ROMA as a marker of disease progression should be further explored in larger studies, especially in premenopausal patients, as it may be superior to CT for the detection of disseminations of less than 2 cm.
Acknowledgements We thank Yuko Ijima and Akiko Iwasa for their excellent technical assistance. We also thank Kenji Abe and Dr. Tetsuo Sekino, EIDIA Co., Ltd., for their support. We appreciate the kind help of Dr. Houda Alachkar for English editing.
Authors’ Contributions Y.Ik. carried out the data collection and drafted the manuscript. K.H. participated in the design of the study. A.K. carried out the statistical analysis. T.H., D.S., and Y.Im. collected the clinical samples. Y.Im. and T.N. carried out the data collection. K.O. participated in the design of the study and reading the proofs of the manuscript. K.F. participated in the design of the study.
Disclosure Statement All the authors declare no financial competing interests.
References
280
4 Fujiwara K, Nagao S, Aotani E, Hasegawa K: Principle and evolving role of intraperitoneal chemotherapy in ovarian cancer. Expert Opin Pharmacother 2013; 14: 1797–1806. 5 Coakley FV, Choi PH, Gougoutas CA, Pothuri B, Venkatraman E, Chi D, Bergman A, Hricak H: Peritoneal metastases: detection with spiral CT in patients with ovarian cancer. Radiology 2002; 223: 495–499. 6 Patel CM, Sahdev A, Reznek RH: CT, MRI and PET imaging in peritoneal malignancy. Cancer Imaging 2011; 11: 123–139. 7 Sehouli J, Senyuva F, Fotopoulou C, Neumann U, Denkert C, Werner L, Gulten OO: Intra-abdominal tumor dissemination pattern and surgical outcome in 214 patients with primary ovarian cancer. J Surg Oncol 2009; 99: 424–427. 8 Davis KR, New PF, Solis OJ, Roberson GH: A review of the findings on computed cranial tomography following intravenous contrast media. Rev Interam Radiol 1977; 2: 15–18.
Oncol Res Treat 2015;38:276–281
9 Partridge E, Kreimer AR, Greenlee RT, Williams C, Xu JL, Church TR, Kessel B, Johnson CC, Weissfeld JL, Isaacs C, Andriole GL, Ogden S, Ragard LR, Buys SS, Team PP: Results from four rounds of ovarian cancer screening in a randomized trial. Obstet Gynecol 2009; 113: 775–782. 10 Hellstrom I, Raycraft J, Hayden-Ledbetter M, Ledbetter JA, Schummer M, McIntosh M, Drescher C, Urban N, Hellstrom KE: The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer Res 2003; 63: 3695–3700. 11 Moore RG, McMeekin DS, Brown AK, DiSilvestro P, Miller MC, Allard WJ, Gajewski W, Kurman R, Bast RC Jr, Skates SJ: A novel multiple marker bioassay utilizing HE4 and CA125 for the prediction of ovarian cancer in patients with a pelvic mass. Gynecol Oncol 2009; 112: 40–46.
Ikeda/Hasegawa/Kurosaki/Miyara/Hanaoka/ Shintani/Imai/Nishikawa/Oda/Fujiwara
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
1 Siegel R, Naishadham D, Jemal A: Cancer statistics, 2013. CA Cancer J Clin 2013; 63: 11–30. 2 Wimberger P, Lehmann N, Kimmig R, Burges A, Meier W, Du Bois A; Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study Group: Prognostic factors for complete debulking in advanced ovarian cancer and its impact on survival. An exploratory analysis of a prospectively randomized phase III study of the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study Group (AGOOVAR). Gynecol Oncol 2007; 106: 69–74. 3 Fotopoulou C, Richter R, Braicu EI, Schmidt SC, Lichtenegger W, Sehouli J: Can complete tumor resection be predicted in advanced primary epithelial ovarian cancer? A systematic evaluation of 360 consecutive patients. Eur J Surg Oncol 2010; 36: 1202–1210.
ROMA as a Peritoneal Dissemination Marker in Ovarian Cancer
16 Gershenson DM: Treatment of ovarian cancer in young women. Clin Obstet Gynecol 2012; 55: 65–74. 17 Fedele L, Vercellini P, Arcaini L, da Dalt MG, Candiani GB: CA 125 in serum, peritoneal fluid, active lesions, and endometrium of patients with endometriosis. Am J Obstet Gynecol 1988; 158: 166–170. 18 Panorchan K, Davenport A: Diagnostic and prognostic role of peritoneal CA 125 in peritoneal dialysis patients presenting with acute peritonitis. BMC Nephrol 2014; 15: 149. 19 Check JH, Vetter BH: A challenge to the concept of tubal reflux to explain the rise and fall of CA125 in serum during the first trimester. Hum Reprod 1995;10: 674–676. 20 Holcomb K, Vucetic Z, Miller MC, Knapp RC: Human epididymis protein 4 offers superior specificity in the differentiation of benign and malignant adnexal masses in premenopausal women. Am J Obstet Gynecol 2011; 205: 358.e1–e6.
21 Anastasi E, Granato T, Marchei GG, Viggiani V, Colaprisca B, Comploj S, Reale MG, Frati L, Midulla C: Ovarian tumor marker HE4 is differently expressed during the phases of the menstrual cycle in healthy young women. Tumour Biol 2010; 31: 411–415. 22 Montagnana M, Danese E, Ruzzenente O, Bresciani V, Nuzzo T, Gelati M, Salvagno GL, Franchi M, Lippi G, Guidi GC: The ROMA (risk of ovarian malignancy algorithm) for estimating the risk of epithelial ovarian cancer in women presenting with pelvic mass: Is it really useful? Clin Chem Lab Med 2011; 49: 521–525. 23 Manegold-Brauer G, Bellin AK, Tercanli S, Lapaire O, Heinzelmann-Schwarz V: The special role of ultrasound for screening, staging and surveillance of malignant ovarian tumors: distinction from other methods of diagnostic imaging. Arch Gynecol Obstet 2014; 289: 491–498.
Oncol Res Treat 2015;38:276–281
281
Downloaded by: NYU Medical Center Library 198.143.38.1 - 8/4/2015 12:34:13 PM
12 Moore RG, Jabre-Raughley M, Brown AK, Robison KM, Miller MC, Allard WJ, Kurman RJ, Bast RC, Skates SJ: Comparison of a novel multiple marker assay vs the risk of malignancy index for the prediction of epithelial ovarian cancer in patients with a pelvic mass. Am J Obstet Gynecol 2010; 203: 228.e1–e6. 13 Li F, Tie R, Chang K, Wang F, Deng S, Lu W, Yu L, Chen M: Does risk for ovarian malignancy algorithm excel human epididymis protein 4 and CA125 in predicting epithelial ovarian cancer: a meta-analysis. BMC Cancer 2012; 12: 258. 14 Karlsen MA, Sandhu N, Hogdall C, Christensen IJ, Nedergaard L, Lundvall L, Engelholm SA, Pedersen AT, Hartwell D, Lydolph M, Laursen IA, Hogdall EV: Evaluation of HE4, CA125, risk of ovarian malignancy algorithm (ROMA) and risk of malignancy index (RMI) as diagnostic tools of epithelial ovarian cancer in patients with a pelvic mass. Gynecol Oncol 2012; 127: 379–383. 15 Sandri MT, Bottari F, Franchi D, Boveri S, Candiani M, Ronzoni S, Peiretti M, Radice D, Passerini R, Sideri M: Comparison of HE4, CA125 and ROMA algorithm in women with a pelvic mass: correlation with pathological outcome. Gynecol Oncol 2013; 128: 233–238.
