G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS European Journal of Radiology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage夽 Ji-Won Oh a , Sung Eun Rha a,∗ , Soon Nam Oh a , Michael Yong Park a , Jae Young Byun a , Ahwon Lee b a Department of Radiology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea b Department of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 4 July 2014 Received in revised form 26 November 2014 Accepted 6 January 2015 Keywords: MRI Diffusion-weighted imaging Apparent diffusion coefficient Epithelial ovarian cancer Histologic grade
a b s t r a c t Objective: The purpose of this article is to correlate the apparent diffusion coefficient (ADC) values of epithelial ovarian cancers with histologic grade and surgical stage. Materials and methods: We enrolled 43 patients with pathologically proven epithelial ovarian cancers for this retrospective study. All patients underwent preoperative pelvic magnetic resonance imaging (MRI) including diffusion-weighted images with b value of 0 and 1000 s/mm2 at 3.0-T unit. The mean ADC values of the solid portion of the tumor were measured and compared among different histologic grades and surgical stages. Results: The mean ADC values of epithelial ovarian cancers differed significantly between grade 1 (welldifferentiated) and grade 2 (moderately-differentiated) (P = 0.013) as well as between grade 1 and grade 3 (poorly-differentiated) (P = 0.01); however, no statistically significant difference existed between grade 2 and grade 3 (P = 0.737). The receiver-operating characteristic analysis indicated that a cutoff ADC value of less than or equal to 1.09 × 10−3 mm2 /s was associated with 94.4% sensitivity and 85.7% specificity in distinguishing grade 1 and grade 2/3 cancer. The difference in mean ADC values was statistically significant for early stage (FIGO stage I) and advanced stage (FIGO stage II-IV) cancer (P = 0.011). The interobserver agreement for the mean ADC values of epithelial ovarian cancers was excellent. Conclusion: The mean ADC values of the solid portion of epithelial ovarian cancers negatively correlated to histologic grade and surgical stage. The mean ADC values may be useful imaging biomarkers for assessment of tumor grade of epithelial ovarian cancer. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Epithelial ovarian cancer is one of the most fatal cancers of the female reproductive tract. Multiple prognostic factors of epithelial ovarian cancer, such as stage of the disease, debulking status after primary surgery, volume of ascites, patient age, performance status, histologic grade and various molecular markers such as her-2/neu, help clinicians evaluate and predict the
夽 The authors did not receive funding for this work. ∗ Corresponding author. Tel.: +82 2 2258 1430; fax: +82 2 599 6771. E-mail addresses: [email protected] (J.-W. Oh), [email protected], [email protected] (S.E. Rha), [email protected] (S.N. Oh), [email protected] (M.Y. Park), [email protected] (J.Y. Byun), [email protected] (A. Lee).
outcome of treatment [1]. Although the strongest independent prognostic factor is the absence of residual tumor after primary surgery [2], the surgical stage and the histologic grade of the tumor, are important prognostic factors in determining the survival [3,4]. The stage of an epithelial ovarian cancer is classified according to the International Federation of Gynecology and Obstetrics (FIGO) staging system, and the histologic grade of the tumor (G1 = well-differentiated, G2 = moderatelydifferentiated, G3 = poorly-differentiated) is determined either by the pattern of differentiation or by the extent of cellular anaplasia as well as the proportion of undifferentiated cells either according to the FIGO system or the World Health Organization (WHO) system [5]. In body imaging, diffusion-weighted imaging (DWI) has become accepted as a useful imaging technique to detect and characterize
http://dx.doi.org/10.1016/j.ejrad.2015.01.005 0720-048X/© 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
2
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
focal lesions [6–10]. Furthermore, the quantitative measurement of apparent diffusion coefficient (ADC) values has also been proposed as a useful imaging biomarker in various types of tumors to differentiate benign and malignant lesions as well as to assess tumor grade and aggressiveness that help predict survival rates [7,11–17]. Recently, DWI has also been increasingly applied as a routine imaging protocol to evaluate the female pelvis due to advances in MR techniques [18,19]. The clinical application of DWI for epithelial ovarian cancer is predominantly concentrated on the differentiation of malignant tissues from benign tissues, detection of peritoneal dissemination and disease stage [20]. Previous studies have demonstrated that almost all invasive malignant tumors showed high signal intensity in the solid component on DWI, although DWI has some limitations to differentiate the benign and malignant solid components of ovarian masses due to false negative results for borderline tumors and false positive results for several benign fibrous tumors. These studies also indicated that the mean ADC values of malignant ovarian surface epithelial tumors were significantly lower than those of benign tumors [21–24]. To the best of our knowledge, no research has evaluated the relationship of ADC values on DWI and prognostic factors for epithelial ovarian cancer such as histologic grade or surgical stage. This study correlates the ADC value of epithelial ovarian cancer on DWI with the histologic grade and surgical stage of the tumor. 2. Materials and methods 2.1. Patient selection Our institutional review board approved this retrospective study and a waiver of informed consent was obtained. The imaging database of our institute was retrospectively queried to identify women who underwent pelvic MR imaging for the preoperative evaluation of complex adnexal masses between March 2009 and June 2013. A total of 441 patients were initially identified. Patients were enrolled in the study if their pelvic MR imaging included DWI with b value of 0 and 1000 s/mm2 at 3.0-T unit and if subsequent surgery was performed for the pathologic confirmation of an invasive epithelial ovarian carcinoma diagnosis. Epithelial origin tumors diagnosed of borderline malignant potential (n = 5) were excluded from the study population, because solid component of the tumor was too small to measure the mean ADC values. Finally, 45 patients with pathologically proven invasive epithelial ovarian carcinomas fulfilled the inclusion criteria. Only the larger mass for bilateral lesions with the same histology (n = 21) was included in the statistical analysis in order to limit an intracorrelation effect. Two patients with clear cell carcinomas were excluded from the analysis, since clear cell carcinomas are usually not histologically graded due to the invariably high-grade nuclei that are identified with this histologic subtype [25]. 2.2. MR imaging technique MRI was performed at 3.0 T (Verio, Siemens Healthcare, Germany) using a pelvic phased-array coil. All patients fasted for 3 h and received an intravenous antispasmodic drug (Buscopan; Boehringer Ingelheim, Korea, Seoul, South Korea) immediately before MRI to reduce bowel peristalsis. Standardized MRI of the pelvis consisted of the following sequence: three-plane localizer; coronal half-Fourier acquisition single-shot turbo spin-echo (HASTE) (TR/TE, 1000 ms/90 ms; slice thickness, 5 mm; matrix, 384 × 214); axial T1-weighted 2D turbo spin-echo (TSE) (TR/TE, 748–800 ms/13 ms; slice thickness, 5 mm; matrix, 384 × 192–230;
echo train length, 3); axial and sagittal T2-weighted 2D TSE (TR/TE, 3950–4600 ms/102 ms; slice thickness, 5 mm; matrix, 384 × 192–230; echo train length, 21); and contrast-enhanced T1weighted 2D TSE (TR/TE, 640 ms/13 ms; slice thickness, 5 mm; matrix, 384 × 192–230; echo train length, 3). Axial DWI was obtained between T2-weighted 2D TSE and contrast-enhanced T1-weighted 2D TSE by using a singleshot spin-echo type echo-planar imaging (EPI) sequence with fat suppression and parallel technique (reduction factor = 2). The DWI imaging parameters were as follows: TR/TE, 10,000 ms/95 ms; b factors, 0, 1000 s/mm2 ; matrix, 100 × 100; field of view, 210–250 × 210–250 mm; and slice thickness, 5 mm; no gap. 2.3. Image analysis MR images were reviewed on a picture archiving and communication system workstation monitor (m-view; Marotech, Seoul, Korea). Conventional MR sequences were evaluated by one radiologist (with 16 years of gynecologic imaging experience), who were blinded to pathologic types, histologic grades and surgical stages. The maximal diameter of the tumors was reported and tumors were classified as predominantly cystic (if less than half of the mass was solid) or predominantly solid (if more half of the mass was solid). The signal intensity of the solid components evaluated on T2-weighted MR images and DWI relative to gluteus muscle and outer myometrium were classified as ‘low’ when the signal was equal to the gluteus muscle, ‘intermediate’ when the signal was equal to the outer myometrium and ‘high’ when higher than the outer myometrium. The quantitative analysis was independently performed by the two radiologists (with 16 and 12 years of gynecologic imaging experience, respectively). A circular ROI was drawn manually on the solid component of the tumor on ADC maps with references of T2weighted images and contrast-enhanced T1-weighted images. ROIs were drawn on a single slice in which the size of the solid portion was the largest. The ROI included as much of the solid portion as possible to obtain the mean ADC values. Meticulous care was taken to avoid necrotic portion, cystic areas, and susceptibility artifacts. ADC values were measured three times and the average value was calculated to get the mean ADC values. 2.4. Analysis of histologic grade and surgical stage The pathologic diagnoses of ovarian surface epithelial carcinomas were made with surgery in all patients. A pathologist blinded to the MRI findings with 15 years of experience in gynecological oncology analyzed the histologic diagnosis and histologic grade of epithelial ovarian carcinomas. Tumors were scored as grade 1 (well-differentiated, G1), grade 2 (moderately-differentiated, G2), or grade 3 (poorly-differentiated, G3) according to microscopic pathologic findings based on the WHO grading system, in which the grade is assigned by the pathologist’s impression of both architectural and cytologic features of the tumor, including the pattern of differentiation, the extent of cellular anaplasia as well as the proportion of undifferentiated cells [5]. Surgical stage classification was based on the FIGO staging [26]. Stage I is a tumor confined to ovaries; stage II consists of pelvic extension of tumor below pelvic brim; stage III includes cytologically or histologically confirmed spread to the peritoneum outside the pelvis and/or metastasis to the retroperitoneal lymph nodes; and stage IV has distant metastasis, such as extra-abdominal metastases including inguinal lymph nodes and lymph nodes outside of the abdominal cavity. We classified stage I cancers as early stage cancers and stage II-IV cancers as advanced stage cancers.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
3
Table 1 Summary of patient characteristics. Characteristics
Data
No of patients Age, mean (range) Tumor size, mean (range) Histologic type Serous Mucinous Endometrioid Transitional cell Mixed serous and transitional cell Histologic grade Well-differentiated (G1) Moderately-differentiated (G2) Poorly-differentiated (G3) Surgical FIGO stage I II III IV
43 53 (29–75) years 9.3 (3–23) cm 32 4 3 2 2 7 27 9 13 3 25 2
Note: Data indicates number of patients unless otherwise indicated.
2.5. Statistical analysis Statistical analysis was performed with the data analysis program SPSS (version 19.0; SPSS Inc., Chicago, IL). A P value less than 0.05 was considered to indicate a statistically significant difference. The agreement between ADC values assigned by the reviewers was analyzed with an intraclass correlation coefficient. A Kolmogorov-Smirnov test evaluated the data distribution normality. The mean ADC values of ovarian cancers were compared among different histologic grades using one-way ANOVA with Scheffe’s post hoc test. An independent sample t-test compared the mean ADC values of early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) cancer. Receiver-operating characteristic (ROC) analysis evaluated the discriminatory capability of ADC values to distinguish G1 and G2/3 carcinomas, as well as early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) carcinomas. ADC values that corresponded to the highest Youden Index (Youden Index = sensitivity + specificity − 1) were chosen as optimal ADC threshold values. In addition, Spearman rank correlation coefficients were calculated to evaluate the correlation between the histologic grade and the surgical stage. 3. Results Table 1 summarizes the characteristics of patients included in the study. Final histopathologic findings of 43 patients (mean age 53 years; range 29–75) revealed 32 serous carcinomas (2 G1, 24 G2, and 6 G3), 4 mucinous carcinomas (3 G1 and 1 G2), 3 endometrioid carcinomas (2 G1 and 1 G2), 2 transitional cell carcinomas (2 G3), and 2 mixed serous and transitional cell carcinomas (1 G2 and 1 G3). The tumor was located in the right adnexa in 14 patients, in the left adnexa in 8, and in both adnexa in 21. The maximal diameter of the tumor ranged from 3 cm to 23 cm (mean, 9.3 cm). A total of 28 tumors showed a predominantly cystic appearance and 15 tumors showed a predominantly solid appearance. All tumors had enhanced solid components. On T2-weighted images, 39 of 43 lesions showed intermediate signal intensity within the solid component and 4 lesions showed high signal intensity compared to the signal intensity of the myometrium. On contrast-enhanced T1-weighted images, 40 of 43 lesions showed a marked contrast enhancement of the solid component and 3 lesions showed a weak contrast enhancement. On DWI, 41 of 43 lesions had high b1000 signal intensity within the solid component and only 2 lesions (G1 mucinous cystadenocarcinomas) showed intermediate signal intensity.
Fig. 1. Box plot comparison of the ADCs among G1, G2 and G3 epithelial ovarian cancer. (A) Reviewer 1. (B) Reviewer 2. Mean ADC values were statistically different between G1 and G2 cancer and between G1 and G3 cancer, but there was some overlap between them; however, there was no statistically significant difference between G2 and G3 cancer.
The interobserver correlation for the mean ADC values of the solid component of epithelial ovarian cancers was excellent ( value = 0.95, P < 0.001). Kolmogorov–Smirnov test results indicated that all subgroups were normally distributed and none of the original variables needed transformations for parametric statistical analysis. The mean ADC values of the solid portion of epithelial ovarian cancers were as follows: G1, 1.23 ± 0.25 (×10−3 mm2 /s); G2, 0.92 ± 0.25; G3, 0.85 ± 0.12 in reviewer 1, and G1, 1.30 ± 0.28; G2, 0.90 ± 0.29; G3, 0.83 ± 0.15 in reviewer 2. A box and whisker plot (Fig. 1) compared the mean ADC values of both reviewers for G1, G2 and G3 cancer. Although there was an overlap of ADC values between different patient groups, both reviewers indicated that the mean ADC values of the solid portion of epithelial ovarian cancers were statistically different for G1 and G2 (P = 0.013 in reviewer 1 and P = 0.003 in reviewer 2) as well as G1 and G3 (P = 0.01 in reviewer 1 and P = 0.003 in reviewer 2), but not for G2 and G3 (P = 0.737 in reviewer 1 and P = 0.741 in reviewer 2). The highest accuracy of 91.7% was obtained with a sensitivity of 94.4% and a specificity of 85.7% when an ADC value of less than or equal to 1.09 × 10−3 mm2 /s was used as the cutoff value to distinguish G1 and G2/3 cancer, based on the results from the reviewer 1.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
4
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
Table 2 Comparison of mean ADC values for early stage and advanced stage ovarian cancer.
Reviewer 1 Reviewer 2
ADC ADC
Early stage (FIGO Stage I)
Advanced stage (FIGO Stages II-IV)
P-value
1.17 ± 0.36 1.20 ± 0.40
0.86 ± 0.15 0.84 ± 0.17
0.011 0.007
Note: Data are mean ± standard deviations (×10−3 mm2 /s).
Table 2 shows the mean ADC values of the solid portion of epithelial ovarian cancers, in early (FIGO stage I) and advanced (FIGO stages II–IV) stage. Both reviewers found a significant difference in the mean ADC values of early stage and advanced stage cancer. The highest accuracy of 81.5% was obtained with a sensitivity of 86.7% and a specificity of 69.2% when an ADC value of less than or equal to 1.03 × 10−3 mm2 /s was used as the cutoff value to distinguish early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) cancer, based on the results from the reviewer 1. There was a statistically significant positive correlation between the histologic grade and the surgical stage (rho = 0.477, p = 0.001) of tumors. Representative images of epithelial ovarian cancers with different histologic grades and surgical stages are demonstrated in Figs. 2 and 3. 4. Discussion Our study demonstrated that ADC values of solid components of epithelial ovarian cancer on DWI negatively correlate to histologic
grade and surgical stage in patients with invasive epithelial ovarian cancer. To our knowledge, this is the first study to correlate ADC values with histologic grade and surgical stage in epithelial ovarian cancer. Histologic grade is associated with the cellularity of the tumor; therefore, the negative correlation between the ADC values and the histologic grade may be explained by the strong negative correlation between the ADC values and tumor cellularity, as previously proven in other studies [27]. There was a statistically significant positive correlation between the histologic grade and the surgical stage of tumors, which may subsequently explain the negative correlation between the ADC values and surgical stage. In our study, the mean ADC values of G1 cancers were higher than G2 and G3 cancers and showed a statistically significant difference; however, there was no significant difference in the ADC values of G2 and G3 cancers by both reviewers. Traditionally, ovarian cancer has been graded as well-, moderately-, and poorly-differentiated; however, a 2-tier histologic grading into low-grade and high-grade has been recently proposed. The 2-tier
Fig. 2. 52-year-old woman with well-differentiated stage I endometrioid ovarian carcinoma. (A) Axial T2-weighted image shows a large mixed cystic and solid mass (asterisk) in the right adnexa. (B) Axial contrast-enhanced T1-weighted image shows a focal enhancing solid component (arrow) of the mass. (C) Corresponding axial DWI (b = 1000 s/mm2 ) shows focal increased signal intensity in the solid component (arrow) of the mass. (D) Corresponding axial ADC map shows hypointense solid component. The mean ADC value of the solid portion is 1.49 × 10−3 mm2 /s.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
5
Fig. 3. 76-year-old woman with poorly differentiated stage III serous ovarian carcinoma. (A) Axial T2-weighted image shows a hypointense solid mass (arrows) in the right adnexa and multiple seeding masses (asterisks) in the pelvic cavity and ascites. U = uterus. (B) Axial contrast-enhanced T1-weighted image shows mild enhancement of the right ovarian mass and peritoneal seeding masses. (C) Corresponding axial DWI (b = 1000 s/mm2 ) shows obviously increased signal intensity in the right ovarian mass (arrow) and peritoneal seeding masses. (D) Corresponding axial ADC map shows obviously decreased signal intensity of the mass (arrow). The mean ADC value of the solid portion is 0.68 × 10−3 mm2 /s.
system classified tumors described as well-differentiated in pathology reports as low-grade, and those described as moderately- or poorly-differentiated as high grade. This binary grading system is suggested as a significant predictor of ovarian serous carcinomas, and has important clinical implications as evidenced by the lack of sensitivity of low grade tumors to the standard cytotoxic chemotherapy that is effective for high grade tumors. [28]. Our results indicated a statistically significant difference in ADC values for the solid components of epithelial ovarian cancer on DWI between G1 and G2/3 cancer and were well-correlated with the binary histologic grading system. The mean ADC values of early stage (FIGO stage I) cancer were significantly higher than that of advanced stage (FIGO stages II–IV) cancer. This result is clinically important due to differences in early and advanced stage cancer management. Cytoreductive surgery to remove the tumor and metastases is the preferred method for the management of advanced stage cancer; however, less aggressive surgery may be considered for early stage (FIGO stage I) cancer. For example, the uterus and the contralateral ovary can be preserved to maintain fertility in women with stage IA cancer [25]. Our results suggest that measuring ADC values on DWI by defining ROI in the solid component of an epithelial ovarian cancer represent a meaningful preoperative evaluation since ADC values showed a significant correlation with histologic grade and surgical stage. Even though the initial treatment of ovarian cancer
is a debulking operation, the measurement of ADC values before surgery can help assess the aggressiveness of the tumor and predict the prognosis of the patient. There are several limitations to this study. First, this study was a retrospective study. Second, it included a small sample size of only 43 patients. Third, we included several different histopathologic subtypes of epithelial ovarian cancers, which may be a significant prognostic factor. However, the preoperative distinction of different histopathologic subtypes of epithelial ovarian cancers with MR imaging is very challenging [29]. Therefore, it is difficult to include only a selected histopathologic subtype at the clinical setting of preoperative evaluation. Fourth, the lack of DWI standardization including the choice of b values may be an important limitation to apply the ADC cut off values at other hospitals. In this study, we assessed ADC monoexponentially using two b values, and did not obtain images with three or more b values to separately assess the effects of perfusion and diffusion. [14,30]. And also, the drawing of ROIs to generate ADC values were not standardized and may create inter and intraobserver agreement problems that can limit the generalization of the results. Fifth, a negative correlation between ADC values and histologic grade or surgical stage was proven; however, overlaps in the ADC values of different patient groups made it impossible to determine an exact range of ADC values according to histologic grade or surgical stage. A prospective study with a larger number of patients using a standardized DWI technique is required
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
6
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
to validate the present results. In addition, the distribution of surgical stage was uneven with a relatively large number of stage I and III tumors and a small number of stage IV tumors; therefore, larger studies are required to investigate the correlation of ADC values with the surgical stage and patient survival. Lastly, we did not include patients with benign ovarian disease as a control group, and there is probably some overlap of ADC values between benign and low grade malignant tumours [21]. But, our study showed that ADC values can be used to predict the histologic grade of epithelial ovarian cancers to some extent. In conclusion, the mean ADC values of solid component of an epithelial ovarian cancer negatively correlated with the tumor grade and surgical stage. Therefore, the mean ADC values may be useful imaging biomarkers for assessment of epithelial ovarian cancer. Conflict of interest None. References [1] Clark TG, Stewart ME, Altman DG, Gabra H, Smyth JF. A prognostic model for ovarian cancer. Br J Cancer 2001;85(7):944–52. [2] Vergote I, Trope CG, Amant F, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med 2010;363(10):943–53. [3] Chan JK, Tian C, Monk BJ, et al. Prognostic factors for high-risk earlystage epithelial ovarian cancer: a Gynecologic Oncology Group study. Cancer 2008;112:2202–10. [4] Park HJ, Nam EJ, Rha SY, et al. A new prognostic index model using meta-analysis in early-stage epithelial ovarian cancer. Gynecol Oncol 2012;126(3):357–63. [5] Silverberg SG. Histopathologic grading of ovarian carcinoma: a review and proposal. Int J Gynecol Pathol 2000;19(1):7–15. [6] Coutinho Jr AC, Krishnaraj A, Pires CE, Bittencourt LK, Guimaraes AR. Pelvic applications of diffusion magnetic resonance images. Magn Reson Imaging Clin N Am 2011;19(1):133–57. [7] Kobayashi S, Koga F, Yoshida S, et al. Diagnostic performance of diffusionweighted magnetic resonance imaging in bladder cancer: potential utility of apparent diffusion coefficient values as a biomarker to predict clinical aggressiveness. Eur Radiol 2011;21(10):2178–86. [8] Kuang F, Ren J, Zhong Q, Liyuan F, Huan Y, Chen Z. The value of apparent diffusion coefficient in the assessment of cervical cancer. Eur Radiol 2013;23(4):1050–8. [9] Thoeny HC, Forstner R, De Keyzer F. Genitourinary applications of diffusionweighted MR imaging in the pelvis. Radiology 2012;263(2):326–42. [10] Vargas HA, Akin O, Franiel T, et al. Diffusion-weighted endorectal MR imaging at 3 T for prostate cancer: tumor detection and assessment of aggressiveness. Radiology 2011;259(3):775–84. [11] Akashi M, Nakahusa Y, Yakabe T, et al. Assessment of aggressiveness of rectal cancer using 3-T MRI: correlation between the apparent diffusion coefficient as a potential imaging biomarker and histologic prognostic factors. Acta Radiol 2013;55(5):524–31.
[12] Akita H, Jinzaki M, Kikuchi E, et al. Preoperative T categorization and prediction of histopathologic grading of urothelial carcinoma in renal pelvis using diffusion-weighted MRI. AJR Am J Roentgenol 2011;197(5):1130–6. [13] Cao K, Gao M, Sun YS, et al. Apparent diffusion coefficient of diffusion weighted MRI in endometrial carcinoma—relationship with local invasiveness. Eur J Radiol 2012;81(8):1926–30. [14] Padhani AR, Liu G, Koh DM, et al. Diffusion-weighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia 2009;11(2):102–25. [15] Takeuchi M, Sasaki S, Ito M, et al. Urinary bladder cancer: diffusion-weighted MR imaging—accuracy for diagnosing T stage and estimating histologic grade. Radiology 2009;251(1):112–21. [16] Verma S, Rajesh A, Morales H, et al. Assessment of aggressiveness of prostate cancer: correlation of apparent diffusion coefficient with histologic grade after radical prostatectomy. AJR Am J Roentgenol 2011;196(2):374–81. [17] Yun TJ, Kim JH, Kim KH, Sohn CH, Park SW. Head and neck squamous cell carcinoma: differentiation of histologic grade with standard- and high-b-value diffusion-weighted MRI. Head Neck 2013;35(5):626–31. [18] Namimoto T, Awai K, Nakaura T, Yanaga Y, Hirai T, Yamashita Y. Role of diffusion-weighted imaging in the diagnosis of gynecological diseases. Eur Radiol 2009;19(3):745–60. [19] Sala E, Rockall A, Rangarajan D, Kubik-Huch RA. The role of dynamic contrastenhanced and diffusion weighted magnetic resonance imaging in the female pelvis. Eur J Radiol 2010;76(3):367–85. [20] Punwani S. Diffusion weighted imaging of female pelvic cancers: concepts and clinical applications. Eur J Radiol 2011;78(1):21–9. [21] Li W, Chu C, Cui Y, Zhang P, Zhu M, Diffusion-weighted. MRI: a useful technique to discriminate benign versus malignant ovarian surface epithelial tumors with solid and cystic components. Abdom Imaging 2012;37(5):897–903. [22] Thomassin-Naggara I, Darai E, Cuenod CA, et al. Contribution of diffusionweighted MR imaging for predicting benignity of complex adnexal masses. Eur Radiol 2009;19(6):1544–52. [23] Thomassin-Naggara I, Toussaint I, Perrot N, et al. Characterization of complex adnexal masses: value of adding perfusion- and diffusion-weighted MR imaging to conventional MR imaging. Radiology 2011;258(3):793–803. [24] Zhang P, Cui Y, Li W, Ren G, Chu C, Wu X. Diagnostic accuracy of diffusionweighted imaging with conventional MR imaging for differentiating complex solid and cystic ovarian tumors at 1.5 T. World J Surg Oncol 2012;10:237, http://dx.doi.org/10.1186/1477-7819-10-237. [25] Berek JS, Natarjan S. Ovarian and fallopian tube cancer. In: Berek JSNE, editor. Berek & Novak’s gynecology. Philadelphia: Lippincott Williams & Wilkins; 2007. p. 1458–547. [26] Prat J, Oncology FCoG. Staging classification for cancer of the ovary, fallopian tube, and peritoneum. Int J Gynaecol Obstet 2014;124(1):1–5. [27] Chen L, Liu M, Bao J, et al. The correlation between apparent diffusion coefficient and tumor cellularity in patients: a meta-analysis. PLoS ONE 2013;8(11):e79008. [28] Hannibal CG, Vang R, Junge J, Kjaerbye-Thygesen A, Kurman RJ, Kjaer SK. A binary histologic grading system for ovarian serous carcinoma is an independent prognostic factor: a population-based study of 4317 women diagnosed in Denmark 1978–2006. Gynecol Oncol 2012;125(3):655–60. [29] Bazot M, Nassar-Slaba J, Thomassin-Naggara I, Cortez A, Uzan S, Darai E. MR imaging compared with intraoperative frozen-section examination for the diagnosis of adnexal tumors; correlation with final histology. Eur Radiol 2006;16(12):2687–99. [30] Nougaret S, Tirumani SH, Addley H, Pandey H, Sala E, Reinhold C. Pearls and pitfalls in MRI of gynecologic malignancy with diffusion-weighted technique. AJR Am J Roentgenol 2013;200(2):261–76.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
ARTICLE IN PRESS European Journal of Radiology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage夽 Ji-Won Oh a , Sung Eun Rha a,∗ , Soon Nam Oh a , Michael Yong Park a , Jae Young Byun a , Ahwon Lee b a Department of Radiology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea b Department of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 4 July 2014 Received in revised form 26 November 2014 Accepted 6 January 2015 Keywords: MRI Diffusion-weighted imaging Apparent diffusion coefficient Epithelial ovarian cancer Histologic grade
a b s t r a c t Objective: The purpose of this article is to correlate the apparent diffusion coefficient (ADC) values of epithelial ovarian cancers with histologic grade and surgical stage. Materials and methods: We enrolled 43 patients with pathologically proven epithelial ovarian cancers for this retrospective study. All patients underwent preoperative pelvic magnetic resonance imaging (MRI) including diffusion-weighted images with b value of 0 and 1000 s/mm2 at 3.0-T unit. The mean ADC values of the solid portion of the tumor were measured and compared among different histologic grades and surgical stages. Results: The mean ADC values of epithelial ovarian cancers differed significantly between grade 1 (welldifferentiated) and grade 2 (moderately-differentiated) (P = 0.013) as well as between grade 1 and grade 3 (poorly-differentiated) (P = 0.01); however, no statistically significant difference existed between grade 2 and grade 3 (P = 0.737). The receiver-operating characteristic analysis indicated that a cutoff ADC value of less than or equal to 1.09 × 10−3 mm2 /s was associated with 94.4% sensitivity and 85.7% specificity in distinguishing grade 1 and grade 2/3 cancer. The difference in mean ADC values was statistically significant for early stage (FIGO stage I) and advanced stage (FIGO stage II-IV) cancer (P = 0.011). The interobserver agreement for the mean ADC values of epithelial ovarian cancers was excellent. Conclusion: The mean ADC values of the solid portion of epithelial ovarian cancers negatively correlated to histologic grade and surgical stage. The mean ADC values may be useful imaging biomarkers for assessment of tumor grade of epithelial ovarian cancer. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Epithelial ovarian cancer is one of the most fatal cancers of the female reproductive tract. Multiple prognostic factors of epithelial ovarian cancer, such as stage of the disease, debulking status after primary surgery, volume of ascites, patient age, performance status, histologic grade and various molecular markers such as her-2/neu, help clinicians evaluate and predict the
夽 The authors did not receive funding for this work. ∗ Corresponding author. Tel.: +82 2 2258 1430; fax: +82 2 599 6771. E-mail addresses: [email protected] (J.-W. Oh), [email protected], [email protected] (S.E. Rha), [email protected] (S.N. Oh), [email protected] (M.Y. Park), [email protected] (J.Y. Byun), [email protected] (A. Lee).
outcome of treatment [1]. Although the strongest independent prognostic factor is the absence of residual tumor after primary surgery [2], the surgical stage and the histologic grade of the tumor, are important prognostic factors in determining the survival [3,4]. The stage of an epithelial ovarian cancer is classified according to the International Federation of Gynecology and Obstetrics (FIGO) staging system, and the histologic grade of the tumor (G1 = well-differentiated, G2 = moderatelydifferentiated, G3 = poorly-differentiated) is determined either by the pattern of differentiation or by the extent of cellular anaplasia as well as the proportion of undifferentiated cells either according to the FIGO system or the World Health Organization (WHO) system [5]. In body imaging, diffusion-weighted imaging (DWI) has become accepted as a useful imaging technique to detect and characterize
http://dx.doi.org/10.1016/j.ejrad.2015.01.005 0720-048X/© 2015 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
2
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
focal lesions [6–10]. Furthermore, the quantitative measurement of apparent diffusion coefficient (ADC) values has also been proposed as a useful imaging biomarker in various types of tumors to differentiate benign and malignant lesions as well as to assess tumor grade and aggressiveness that help predict survival rates [7,11–17]. Recently, DWI has also been increasingly applied as a routine imaging protocol to evaluate the female pelvis due to advances in MR techniques [18,19]. The clinical application of DWI for epithelial ovarian cancer is predominantly concentrated on the differentiation of malignant tissues from benign tissues, detection of peritoneal dissemination and disease stage [20]. Previous studies have demonstrated that almost all invasive malignant tumors showed high signal intensity in the solid component on DWI, although DWI has some limitations to differentiate the benign and malignant solid components of ovarian masses due to false negative results for borderline tumors and false positive results for several benign fibrous tumors. These studies also indicated that the mean ADC values of malignant ovarian surface epithelial tumors were significantly lower than those of benign tumors [21–24]. To the best of our knowledge, no research has evaluated the relationship of ADC values on DWI and prognostic factors for epithelial ovarian cancer such as histologic grade or surgical stage. This study correlates the ADC value of epithelial ovarian cancer on DWI with the histologic grade and surgical stage of the tumor. 2. Materials and methods 2.1. Patient selection Our institutional review board approved this retrospective study and a waiver of informed consent was obtained. The imaging database of our institute was retrospectively queried to identify women who underwent pelvic MR imaging for the preoperative evaluation of complex adnexal masses between March 2009 and June 2013. A total of 441 patients were initially identified. Patients were enrolled in the study if their pelvic MR imaging included DWI with b value of 0 and 1000 s/mm2 at 3.0-T unit and if subsequent surgery was performed for the pathologic confirmation of an invasive epithelial ovarian carcinoma diagnosis. Epithelial origin tumors diagnosed of borderline malignant potential (n = 5) were excluded from the study population, because solid component of the tumor was too small to measure the mean ADC values. Finally, 45 patients with pathologically proven invasive epithelial ovarian carcinomas fulfilled the inclusion criteria. Only the larger mass for bilateral lesions with the same histology (n = 21) was included in the statistical analysis in order to limit an intracorrelation effect. Two patients with clear cell carcinomas were excluded from the analysis, since clear cell carcinomas are usually not histologically graded due to the invariably high-grade nuclei that are identified with this histologic subtype [25]. 2.2. MR imaging technique MRI was performed at 3.0 T (Verio, Siemens Healthcare, Germany) using a pelvic phased-array coil. All patients fasted for 3 h and received an intravenous antispasmodic drug (Buscopan; Boehringer Ingelheim, Korea, Seoul, South Korea) immediately before MRI to reduce bowel peristalsis. Standardized MRI of the pelvis consisted of the following sequence: three-plane localizer; coronal half-Fourier acquisition single-shot turbo spin-echo (HASTE) (TR/TE, 1000 ms/90 ms; slice thickness, 5 mm; matrix, 384 × 214); axial T1-weighted 2D turbo spin-echo (TSE) (TR/TE, 748–800 ms/13 ms; slice thickness, 5 mm; matrix, 384 × 192–230;
echo train length, 3); axial and sagittal T2-weighted 2D TSE (TR/TE, 3950–4600 ms/102 ms; slice thickness, 5 mm; matrix, 384 × 192–230; echo train length, 21); and contrast-enhanced T1weighted 2D TSE (TR/TE, 640 ms/13 ms; slice thickness, 5 mm; matrix, 384 × 192–230; echo train length, 3). Axial DWI was obtained between T2-weighted 2D TSE and contrast-enhanced T1-weighted 2D TSE by using a singleshot spin-echo type echo-planar imaging (EPI) sequence with fat suppression and parallel technique (reduction factor = 2). The DWI imaging parameters were as follows: TR/TE, 10,000 ms/95 ms; b factors, 0, 1000 s/mm2 ; matrix, 100 × 100; field of view, 210–250 × 210–250 mm; and slice thickness, 5 mm; no gap. 2.3. Image analysis MR images were reviewed on a picture archiving and communication system workstation monitor (m-view; Marotech, Seoul, Korea). Conventional MR sequences were evaluated by one radiologist (with 16 years of gynecologic imaging experience), who were blinded to pathologic types, histologic grades and surgical stages. The maximal diameter of the tumors was reported and tumors were classified as predominantly cystic (if less than half of the mass was solid) or predominantly solid (if more half of the mass was solid). The signal intensity of the solid components evaluated on T2-weighted MR images and DWI relative to gluteus muscle and outer myometrium were classified as ‘low’ when the signal was equal to the gluteus muscle, ‘intermediate’ when the signal was equal to the outer myometrium and ‘high’ when higher than the outer myometrium. The quantitative analysis was independently performed by the two radiologists (with 16 and 12 years of gynecologic imaging experience, respectively). A circular ROI was drawn manually on the solid component of the tumor on ADC maps with references of T2weighted images and contrast-enhanced T1-weighted images. ROIs were drawn on a single slice in which the size of the solid portion was the largest. The ROI included as much of the solid portion as possible to obtain the mean ADC values. Meticulous care was taken to avoid necrotic portion, cystic areas, and susceptibility artifacts. ADC values were measured three times and the average value was calculated to get the mean ADC values. 2.4. Analysis of histologic grade and surgical stage The pathologic diagnoses of ovarian surface epithelial carcinomas were made with surgery in all patients. A pathologist blinded to the MRI findings with 15 years of experience in gynecological oncology analyzed the histologic diagnosis and histologic grade of epithelial ovarian carcinomas. Tumors were scored as grade 1 (well-differentiated, G1), grade 2 (moderately-differentiated, G2), or grade 3 (poorly-differentiated, G3) according to microscopic pathologic findings based on the WHO grading system, in which the grade is assigned by the pathologist’s impression of both architectural and cytologic features of the tumor, including the pattern of differentiation, the extent of cellular anaplasia as well as the proportion of undifferentiated cells [5]. Surgical stage classification was based on the FIGO staging [26]. Stage I is a tumor confined to ovaries; stage II consists of pelvic extension of tumor below pelvic brim; stage III includes cytologically or histologically confirmed spread to the peritoneum outside the pelvis and/or metastasis to the retroperitoneal lymph nodes; and stage IV has distant metastasis, such as extra-abdominal metastases including inguinal lymph nodes and lymph nodes outside of the abdominal cavity. We classified stage I cancers as early stage cancers and stage II-IV cancers as advanced stage cancers.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
3
Table 1 Summary of patient characteristics. Characteristics
Data
No of patients Age, mean (range) Tumor size, mean (range) Histologic type Serous Mucinous Endometrioid Transitional cell Mixed serous and transitional cell Histologic grade Well-differentiated (G1) Moderately-differentiated (G2) Poorly-differentiated (G3) Surgical FIGO stage I II III IV
43 53 (29–75) years 9.3 (3–23) cm 32 4 3 2 2 7 27 9 13 3 25 2
Note: Data indicates number of patients unless otherwise indicated.
2.5. Statistical analysis Statistical analysis was performed with the data analysis program SPSS (version 19.0; SPSS Inc., Chicago, IL). A P value less than 0.05 was considered to indicate a statistically significant difference. The agreement between ADC values assigned by the reviewers was analyzed with an intraclass correlation coefficient. A Kolmogorov-Smirnov test evaluated the data distribution normality. The mean ADC values of ovarian cancers were compared among different histologic grades using one-way ANOVA with Scheffe’s post hoc test. An independent sample t-test compared the mean ADC values of early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) cancer. Receiver-operating characteristic (ROC) analysis evaluated the discriminatory capability of ADC values to distinguish G1 and G2/3 carcinomas, as well as early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) carcinomas. ADC values that corresponded to the highest Youden Index (Youden Index = sensitivity + specificity − 1) were chosen as optimal ADC threshold values. In addition, Spearman rank correlation coefficients were calculated to evaluate the correlation between the histologic grade and the surgical stage. 3. Results Table 1 summarizes the characteristics of patients included in the study. Final histopathologic findings of 43 patients (mean age 53 years; range 29–75) revealed 32 serous carcinomas (2 G1, 24 G2, and 6 G3), 4 mucinous carcinomas (3 G1 and 1 G2), 3 endometrioid carcinomas (2 G1 and 1 G2), 2 transitional cell carcinomas (2 G3), and 2 mixed serous and transitional cell carcinomas (1 G2 and 1 G3). The tumor was located in the right adnexa in 14 patients, in the left adnexa in 8, and in both adnexa in 21. The maximal diameter of the tumor ranged from 3 cm to 23 cm (mean, 9.3 cm). A total of 28 tumors showed a predominantly cystic appearance and 15 tumors showed a predominantly solid appearance. All tumors had enhanced solid components. On T2-weighted images, 39 of 43 lesions showed intermediate signal intensity within the solid component and 4 lesions showed high signal intensity compared to the signal intensity of the myometrium. On contrast-enhanced T1-weighted images, 40 of 43 lesions showed a marked contrast enhancement of the solid component and 3 lesions showed a weak contrast enhancement. On DWI, 41 of 43 lesions had high b1000 signal intensity within the solid component and only 2 lesions (G1 mucinous cystadenocarcinomas) showed intermediate signal intensity.
Fig. 1. Box plot comparison of the ADCs among G1, G2 and G3 epithelial ovarian cancer. (A) Reviewer 1. (B) Reviewer 2. Mean ADC values were statistically different between G1 and G2 cancer and between G1 and G3 cancer, but there was some overlap between them; however, there was no statistically significant difference between G2 and G3 cancer.
The interobserver correlation for the mean ADC values of the solid component of epithelial ovarian cancers was excellent ( value = 0.95, P < 0.001). Kolmogorov–Smirnov test results indicated that all subgroups were normally distributed and none of the original variables needed transformations for parametric statistical analysis. The mean ADC values of the solid portion of epithelial ovarian cancers were as follows: G1, 1.23 ± 0.25 (×10−3 mm2 /s); G2, 0.92 ± 0.25; G3, 0.85 ± 0.12 in reviewer 1, and G1, 1.30 ± 0.28; G2, 0.90 ± 0.29; G3, 0.83 ± 0.15 in reviewer 2. A box and whisker plot (Fig. 1) compared the mean ADC values of both reviewers for G1, G2 and G3 cancer. Although there was an overlap of ADC values between different patient groups, both reviewers indicated that the mean ADC values of the solid portion of epithelial ovarian cancers were statistically different for G1 and G2 (P = 0.013 in reviewer 1 and P = 0.003 in reviewer 2) as well as G1 and G3 (P = 0.01 in reviewer 1 and P = 0.003 in reviewer 2), but not for G2 and G3 (P = 0.737 in reviewer 1 and P = 0.741 in reviewer 2). The highest accuracy of 91.7% was obtained with a sensitivity of 94.4% and a specificity of 85.7% when an ADC value of less than or equal to 1.09 × 10−3 mm2 /s was used as the cutoff value to distinguish G1 and G2/3 cancer, based on the results from the reviewer 1.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
4
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
Table 2 Comparison of mean ADC values for early stage and advanced stage ovarian cancer.
Reviewer 1 Reviewer 2
ADC ADC
Early stage (FIGO Stage I)
Advanced stage (FIGO Stages II-IV)
P-value
1.17 ± 0.36 1.20 ± 0.40
0.86 ± 0.15 0.84 ± 0.17
0.011 0.007
Note: Data are mean ± standard deviations (×10−3 mm2 /s).
Table 2 shows the mean ADC values of the solid portion of epithelial ovarian cancers, in early (FIGO stage I) and advanced (FIGO stages II–IV) stage. Both reviewers found a significant difference in the mean ADC values of early stage and advanced stage cancer. The highest accuracy of 81.5% was obtained with a sensitivity of 86.7% and a specificity of 69.2% when an ADC value of less than or equal to 1.03 × 10−3 mm2 /s was used as the cutoff value to distinguish early stage (FIGO stage I) and advanced stage (FIGO stages II–IV) cancer, based on the results from the reviewer 1. There was a statistically significant positive correlation between the histologic grade and the surgical stage (rho = 0.477, p = 0.001) of tumors. Representative images of epithelial ovarian cancers with different histologic grades and surgical stages are demonstrated in Figs. 2 and 3. 4. Discussion Our study demonstrated that ADC values of solid components of epithelial ovarian cancer on DWI negatively correlate to histologic
grade and surgical stage in patients with invasive epithelial ovarian cancer. To our knowledge, this is the first study to correlate ADC values with histologic grade and surgical stage in epithelial ovarian cancer. Histologic grade is associated with the cellularity of the tumor; therefore, the negative correlation between the ADC values and the histologic grade may be explained by the strong negative correlation between the ADC values and tumor cellularity, as previously proven in other studies [27]. There was a statistically significant positive correlation between the histologic grade and the surgical stage of tumors, which may subsequently explain the negative correlation between the ADC values and surgical stage. In our study, the mean ADC values of G1 cancers were higher than G2 and G3 cancers and showed a statistically significant difference; however, there was no significant difference in the ADC values of G2 and G3 cancers by both reviewers. Traditionally, ovarian cancer has been graded as well-, moderately-, and poorly-differentiated; however, a 2-tier histologic grading into low-grade and high-grade has been recently proposed. The 2-tier
Fig. 2. 52-year-old woman with well-differentiated stage I endometrioid ovarian carcinoma. (A) Axial T2-weighted image shows a large mixed cystic and solid mass (asterisk) in the right adnexa. (B) Axial contrast-enhanced T1-weighted image shows a focal enhancing solid component (arrow) of the mass. (C) Corresponding axial DWI (b = 1000 s/mm2 ) shows focal increased signal intensity in the solid component (arrow) of the mass. (D) Corresponding axial ADC map shows hypointense solid component. The mean ADC value of the solid portion is 1.49 × 10−3 mm2 /s.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
5
Fig. 3. 76-year-old woman with poorly differentiated stage III serous ovarian carcinoma. (A) Axial T2-weighted image shows a hypointense solid mass (arrows) in the right adnexa and multiple seeding masses (asterisks) in the pelvic cavity and ascites. U = uterus. (B) Axial contrast-enhanced T1-weighted image shows mild enhancement of the right ovarian mass and peritoneal seeding masses. (C) Corresponding axial DWI (b = 1000 s/mm2 ) shows obviously increased signal intensity in the right ovarian mass (arrow) and peritoneal seeding masses. (D) Corresponding axial ADC map shows obviously decreased signal intensity of the mass (arrow). The mean ADC value of the solid portion is 0.68 × 10−3 mm2 /s.
system classified tumors described as well-differentiated in pathology reports as low-grade, and those described as moderately- or poorly-differentiated as high grade. This binary grading system is suggested as a significant predictor of ovarian serous carcinomas, and has important clinical implications as evidenced by the lack of sensitivity of low grade tumors to the standard cytotoxic chemotherapy that is effective for high grade tumors. [28]. Our results indicated a statistically significant difference in ADC values for the solid components of epithelial ovarian cancer on DWI between G1 and G2/3 cancer and were well-correlated with the binary histologic grading system. The mean ADC values of early stage (FIGO stage I) cancer were significantly higher than that of advanced stage (FIGO stages II–IV) cancer. This result is clinically important due to differences in early and advanced stage cancer management. Cytoreductive surgery to remove the tumor and metastases is the preferred method for the management of advanced stage cancer; however, less aggressive surgery may be considered for early stage (FIGO stage I) cancer. For example, the uterus and the contralateral ovary can be preserved to maintain fertility in women with stage IA cancer [25]. Our results suggest that measuring ADC values on DWI by defining ROI in the solid component of an epithelial ovarian cancer represent a meaningful preoperative evaluation since ADC values showed a significant correlation with histologic grade and surgical stage. Even though the initial treatment of ovarian cancer
is a debulking operation, the measurement of ADC values before surgery can help assess the aggressiveness of the tumor and predict the prognosis of the patient. There are several limitations to this study. First, this study was a retrospective study. Second, it included a small sample size of only 43 patients. Third, we included several different histopathologic subtypes of epithelial ovarian cancers, which may be a significant prognostic factor. However, the preoperative distinction of different histopathologic subtypes of epithelial ovarian cancers with MR imaging is very challenging [29]. Therefore, it is difficult to include only a selected histopathologic subtype at the clinical setting of preoperative evaluation. Fourth, the lack of DWI standardization including the choice of b values may be an important limitation to apply the ADC cut off values at other hospitals. In this study, we assessed ADC monoexponentially using two b values, and did not obtain images with three or more b values to separately assess the effects of perfusion and diffusion. [14,30]. And also, the drawing of ROIs to generate ADC values were not standardized and may create inter and intraobserver agreement problems that can limit the generalization of the results. Fifth, a negative correlation between ADC values and histologic grade or surgical stage was proven; however, overlaps in the ADC values of different patient groups made it impossible to determine an exact range of ADC values according to histologic grade or surgical stage. A prospective study with a larger number of patients using a standardized DWI technique is required
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
G Model EURR-7007; No. of Pages 6
ARTICLE IN PRESS
6
J.-W. Oh et al. / European Journal of Radiology xxx (2015) xxx–xxx
to validate the present results. In addition, the distribution of surgical stage was uneven with a relatively large number of stage I and III tumors and a small number of stage IV tumors; therefore, larger studies are required to investigate the correlation of ADC values with the surgical stage and patient survival. Lastly, we did not include patients with benign ovarian disease as a control group, and there is probably some overlap of ADC values between benign and low grade malignant tumours [21]. But, our study showed that ADC values can be used to predict the histologic grade of epithelial ovarian cancers to some extent. In conclusion, the mean ADC values of solid component of an epithelial ovarian cancer negatively correlated with the tumor grade and surgical stage. Therefore, the mean ADC values may be useful imaging biomarkers for assessment of epithelial ovarian cancer. Conflict of interest None. References [1] Clark TG, Stewart ME, Altman DG, Gabra H, Smyth JF. A prognostic model for ovarian cancer. Br J Cancer 2001;85(7):944–52. [2] Vergote I, Trope CG, Amant F, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med 2010;363(10):943–53. [3] Chan JK, Tian C, Monk BJ, et al. Prognostic factors for high-risk earlystage epithelial ovarian cancer: a Gynecologic Oncology Group study. Cancer 2008;112:2202–10. [4] Park HJ, Nam EJ, Rha SY, et al. A new prognostic index model using meta-analysis in early-stage epithelial ovarian cancer. Gynecol Oncol 2012;126(3):357–63. [5] Silverberg SG. Histopathologic grading of ovarian carcinoma: a review and proposal. Int J Gynecol Pathol 2000;19(1):7–15. [6] Coutinho Jr AC, Krishnaraj A, Pires CE, Bittencourt LK, Guimaraes AR. Pelvic applications of diffusion magnetic resonance images. Magn Reson Imaging Clin N Am 2011;19(1):133–57. [7] Kobayashi S, Koga F, Yoshida S, et al. Diagnostic performance of diffusionweighted magnetic resonance imaging in bladder cancer: potential utility of apparent diffusion coefficient values as a biomarker to predict clinical aggressiveness. Eur Radiol 2011;21(10):2178–86. [8] Kuang F, Ren J, Zhong Q, Liyuan F, Huan Y, Chen Z. The value of apparent diffusion coefficient in the assessment of cervical cancer. Eur Radiol 2013;23(4):1050–8. [9] Thoeny HC, Forstner R, De Keyzer F. Genitourinary applications of diffusionweighted MR imaging in the pelvis. Radiology 2012;263(2):326–42. [10] Vargas HA, Akin O, Franiel T, et al. Diffusion-weighted endorectal MR imaging at 3 T for prostate cancer: tumor detection and assessment of aggressiveness. Radiology 2011;259(3):775–84. [11] Akashi M, Nakahusa Y, Yakabe T, et al. Assessment of aggressiveness of rectal cancer using 3-T MRI: correlation between the apparent diffusion coefficient as a potential imaging biomarker and histologic prognostic factors. Acta Radiol 2013;55(5):524–31.
[12] Akita H, Jinzaki M, Kikuchi E, et al. Preoperative T categorization and prediction of histopathologic grading of urothelial carcinoma in renal pelvis using diffusion-weighted MRI. AJR Am J Roentgenol 2011;197(5):1130–6. [13] Cao K, Gao M, Sun YS, et al. Apparent diffusion coefficient of diffusion weighted MRI in endometrial carcinoma—relationship with local invasiveness. Eur J Radiol 2012;81(8):1926–30. [14] Padhani AR, Liu G, Koh DM, et al. Diffusion-weighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia 2009;11(2):102–25. [15] Takeuchi M, Sasaki S, Ito M, et al. Urinary bladder cancer: diffusion-weighted MR imaging—accuracy for diagnosing T stage and estimating histologic grade. Radiology 2009;251(1):112–21. [16] Verma S, Rajesh A, Morales H, et al. Assessment of aggressiveness of prostate cancer: correlation of apparent diffusion coefficient with histologic grade after radical prostatectomy. AJR Am J Roentgenol 2011;196(2):374–81. [17] Yun TJ, Kim JH, Kim KH, Sohn CH, Park SW. Head and neck squamous cell carcinoma: differentiation of histologic grade with standard- and high-b-value diffusion-weighted MRI. Head Neck 2013;35(5):626–31. [18] Namimoto T, Awai K, Nakaura T, Yanaga Y, Hirai T, Yamashita Y. Role of diffusion-weighted imaging in the diagnosis of gynecological diseases. Eur Radiol 2009;19(3):745–60. [19] Sala E, Rockall A, Rangarajan D, Kubik-Huch RA. The role of dynamic contrastenhanced and diffusion weighted magnetic resonance imaging in the female pelvis. Eur J Radiol 2010;76(3):367–85. [20] Punwani S. Diffusion weighted imaging of female pelvic cancers: concepts and clinical applications. Eur J Radiol 2011;78(1):21–9. [21] Li W, Chu C, Cui Y, Zhang P, Zhu M, Diffusion-weighted. MRI: a useful technique to discriminate benign versus malignant ovarian surface epithelial tumors with solid and cystic components. Abdom Imaging 2012;37(5):897–903. [22] Thomassin-Naggara I, Darai E, Cuenod CA, et al. Contribution of diffusionweighted MR imaging for predicting benignity of complex adnexal masses. Eur Radiol 2009;19(6):1544–52. [23] Thomassin-Naggara I, Toussaint I, Perrot N, et al. Characterization of complex adnexal masses: value of adding perfusion- and diffusion-weighted MR imaging to conventional MR imaging. Radiology 2011;258(3):793–803. [24] Zhang P, Cui Y, Li W, Ren G, Chu C, Wu X. Diagnostic accuracy of diffusionweighted imaging with conventional MR imaging for differentiating complex solid and cystic ovarian tumors at 1.5 T. World J Surg Oncol 2012;10:237, http://dx.doi.org/10.1186/1477-7819-10-237. [25] Berek JS, Natarjan S. Ovarian and fallopian tube cancer. In: Berek JSNE, editor. Berek & Novak’s gynecology. Philadelphia: Lippincott Williams & Wilkins; 2007. p. 1458–547. [26] Prat J, Oncology FCoG. Staging classification for cancer of the ovary, fallopian tube, and peritoneum. Int J Gynaecol Obstet 2014;124(1):1–5. [27] Chen L, Liu M, Bao J, et al. The correlation between apparent diffusion coefficient and tumor cellularity in patients: a meta-analysis. PLoS ONE 2013;8(11):e79008. [28] Hannibal CG, Vang R, Junge J, Kjaerbye-Thygesen A, Kurman RJ, Kjaer SK. A binary histologic grading system for ovarian serous carcinoma is an independent prognostic factor: a population-based study of 4317 women diagnosed in Denmark 1978–2006. Gynecol Oncol 2012;125(3):655–60. [29] Bazot M, Nassar-Slaba J, Thomassin-Naggara I, Cortez A, Uzan S, Darai E. MR imaging compared with intraoperative frozen-section examination for the diagnosis of adnexal tumors; correlation with final histology. Eur Radiol 2006;16(12):2687–99. [30] Nougaret S, Tirumani SH, Addley H, Pandey H, Sala E, Reinhold C. Pearls and pitfalls in MRI of gynecologic malignancy with diffusion-weighted technique. AJR Am J Roentgenol 2013;200(2):261–76.
Please cite this article in press as: Oh J-W, et al. Diffusion-weighted MRI of epithelial ovarian cancers: Correlation of apparent diffusion coefficient values with histologic grade and surgical stage. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.005
