ORIGINAL STUDY
Clinical Application of Diffusion-Weighted Magnetic Resonance Imaging in Uterine Cervical Cancer Ying Liu, MD,* Zhaoxiang Ye, MD,* Haoran Sun, MD,Þ and Renju Bai, MDÞ
Objective: This study aimed to investigate the application value of apparent diffusion coefficient (ADC) values in evaluating histological type as well as pathologic grade of uterine cervical cancer; and to investigate whether ADC values could reflect tumor cellular density. Methods: Ninety-eight patients with histopathologically proven uterine cervical cancer were included in this study. Mean ADC value and minimum ADC value of the tumor were measured. Tumor cellular density was counted using colored multifunction imaging analyzing system. Results: Both mean ADC value and minimum ADC value of squamous cell carcinoma were significantly lower than that of adenocarcinoma (P = 0.001; P = 0.000). Using mean ADC criteria (e0.965 10j3 mm2/s) and minimum ADC criteria (e0.844 10j3 mm2/s), the sensitivity and specificity for differentiating squamous cell carcinoma from adenocarcinoma were 83.5% and 76.9%, and 77.6% and 92.3%, respectively. Receiver operating characteristic analysis revealed that there was no statistically significant difference in the Az values between them (P = 0.990). Tumor cellular density, mean ADC value, and minimum ADC value of different pathological grade varied significantly (P = 0.000, P = 0.000, P = 0.000). There was a significant positive linear correlation between tumor cellular density and pathological grade of tumor (P = 0.000). Both mean ADC value and minimum ADC value correlated negatively with cellular density (P = 0.000, P = 0.000) and the pathological grade of tumor (P = 0.000, P = 0.000). Comparisons of correlation coefficients showed no significant differences (P = 0.656, P = 0.631). Conclusions: Diffusion-weighted magnetic resonance imaging has a potential ability to indicate the histologic type of uterine cervical cancer. Apparent diffusion coefficient measurements of uterine cervical cancer can represent tumor cellular density, thus providing a new method for evaluating the pathological grade of tumor. Key Words: Uterine cervical cancer, Diffusion-weighted MR imaging, Cellularity density, Pathological grade Received August 13, 2014, and in revised form March 2, 2015. Accepted for publication March 19, 2015. (Int J Gynecol Cancer 2015;25: 1073Y1078)
*Department of Radiology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center of Cancer, Key Laboratory of Cancer Prevention and Therapy; and †Department of Radiology, General Hospital of Tianjin Medical University, Tianjin, China. Copyright * 2015 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000472 International Journal of Gynecological Cancer
Address correspondence and reprint requests to Ying Liu, MD, Department of Radiology, Tianjin Medical University Cancer Institute and Hospital, Huan-Hu-Xi Rd, Ti-Yuan-Bei, He Xi District, Tianjin 300060, China. E-mail: [email protected]. The authors declare no conflicts of interest. Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net).
& Volume 25, Number 6, July 2015
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1073
International Journal of Gynecological Cancer
Liu et al
cervical cancer is one of the most common gyneU terine cological tumors diagnosed in women worldwide, and is 1
the second most common cause of cancer-related deaths in women living in developing countries.2 Despite major advances in screening and prevention, the relative proportion of cervical adenocarcinoma (AC) and adenosquamous carcinoma (ASC), as compared with squamous cell carcinoma (SCC), has increased.3Y6 Some studies have reported that AC and poorly differentiated tumors show increased likelihood of metastasis and poorer outcomes than their squamous and well-differentiated counterparts, respectively.7Y9 Diffusion-weighted magnetic resonance (MR) imaging (DWI) is a noninvasive technique that supplies information of water proton mobility. It provides information on extracellular space tortuosity, tissue cellularity, and cellular membrane integrity, which enables the characterization of tissue at microscopic level and makes subtle abnormalities more obvious.10Y12 It is being incorporated into oncological imaging practice as a possible tool for characterizing tissue (ie, discriminating between benign and malignant lesions, distinguishing among tumor types and assessing tumor grade, monitoring response to therapy, and detecting recurrent cancer).13Y16 To date, DWI has become a routine imaging protocol for the female pelvis, because it is so quick to perform, and can be simply adopted on most current MRI scanners without any additional new equipment and intravenous contrast agents. Clinical applications of DWI for uterine cervical cancer predominantly focused on tumor detection, staging of disease, differentiating between metastatic and nonYmetastatic lymph nodes, and treatment evaluation.17,18 However, little information is available for capturing some histopathological features of uterine cervical cancer by using DWI. In previous studies,18,19 it has been confirmed that mean apparent diffusion coefficient (ADC) value of uterine cervical cancer could indicate the histologic type and represent tumor cellular density to some extent, however, they did not compare the diagnostic performance between mean ADC value and minimum ADC value. Thus, the purpose of this study was to evaluate whether ADC value could reflect the histological type and tumor differentiation with large samples, meanwhile, investigate the relationship between ADC value and tumor cellular density.
MATERIALS AND METHODS Patient Population Our study received institutional ethics committee approval and informed consent was obtained from all patients. Inclusion criteria consisted of (1) International Federation of Gynecology and Obstetrics stage based on clinical examination ranges from II to IV, (2) no previous radiation or concurrent chemoradiation therapy for uterine cervical cancer before MRI examination. Exclusion criteria consisted of (1) unable to get histologically (biopsy or surgery) proven of uterine cervical cancer, (2) receiving antineoplastic therapy after MRI examination and then undergo surgical operation or biopsy. Ninety-eight female patients (mean age, 49 years; age range, 25Y80 years) with uterine cervical cancer enrolled in this study. Uterine cervical cancer was confirmed in all 98 patients, either by surgery (37 cases) or by biopsy (61 cases).
1074
& Volume 25, Number 6, July 2015
MR Examination Magnetic resonance examinations were performed before surgery and the initiation of concurrent chemoradiation. To limit bowel motion, patients were recommended to fast for 4 to 6 hours before imaging. All MR examinations were performed using a 1.5-T unit (Twin Excite, GE Healthcare, USA) with Torso phased-array body coil. Before DWI, conventional T2-weighted fast spin-echo in the sagittal and transverse planes (TR/TE, 4000 ms/85 ms; matrix size, 320 224; band width, 31.25 Hz/pixel; field of view, 36 cm; number of excitation, 2; slice thickness, 6 mm; gap, 1 mm), T2-weighted fast spin-echo with fat suppression in the transverse plane (the parameters were the same as T2-weighted image) and T1-weighted spin-echo in transverse plane (TR/TE, 500 ms/ 20 ms; matrix size, 320 160; band width, 31.25 Hz/pixel; field of view, 36 cm; number of excitation, 2; slice thickness, 6 mm; gap, 1 mm) were obtained. Transverse diffusion-weighted images were obtained using a nonYbreath-hold single-shot spin-echo echo-planar imaging sequence and array spatial sensitivity encoding technique with the following parameters: repetition time of 4000 milliseconds; echo time of 58.5 milliseconds; matrix size, 128 128; field of view, 36 cm; number of excitation, 4; slice thickness, 6 mm; gap, 1 mm; R factor, 2; phase-encoding direction, anteroposterior; and b value of 0 and 1000 s/mm2. The diffusion-weighting gradients were applied in all 3 orthogonal directions. The scanning time of DWI was 1 minute and 4 seconds.
MR Image Analysis Magnetic resonance images were analyzed in consensus by 2 radiologists. Both observers were blinded to each other’s MR assessments and clinical information including histological type and pathological grade of tumor. Mean ADC value and minimum ADC value were extracted from the manual placement of the region of interest (ROI) on the ADC map. The central slice which could manifest the largest part of tumor was selected. Mean ADC was obtained by a single measurement covering the solid portion of tumor as large as possible, avoiding volume averaging with cystic or degenerative regions that might influence the quantitative data; whereas a minimum ADC was extracted from the manual placement of 5 circular ROIs in the former ROI, and the ROI with the lowest ADC value was selected as the minimum ADC value.
Histopathological Evaluation Ninety-eight cases of uterine cervical cancer were analyzed for histopathological evaluation. A single experienced pathologist with 15 years of experience, who was blinded to MR results, coded and evaluated all slides. Formalin-fixed paraffin-embedded tissue blocks of cervical biopsy specimens taken pretreatment or of resection specimens taken after surgery, were sectioned at a thickness of 4 Km, and then processed for conventional histological assessment by hematoxylin and eosin staining to verify the morphological diagnosis of uterine cervical cancer and to evaluate its differentiation. The differentiation of uterine cervical cancer was classified into well (grade I), moderate (grade II), and poor * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 25, Number 6, July 2015
(grade III). When different tumor grades coexisted within a tumor, the more predominant differentiation of the tumor was selected. For each specimen (original magnification, 200), 5 visual fields that showed representative pathological findings in each carcinoma were chosen by 1 pathologist. Cellular density of each specimen was measured in colored multifunction imaging analyzing system. Tumor cellular density was defined as ratios of the total area of tumor cell nuclei divided by the area of sample image in each specimen, expressed by percentage.
Statistical Analysis Statistical analyses were performed by using statistical software (SPSS, version 13.0) with the exception of the comparison of areas under independent ROC curves and correlation coefficients, and the assessment of interobserver agreement, which were performed using MedCalc for windows (version 11.3.8.0; MedCalc software). The intraclass correlation coefficient (ICC) was used to assess the interobserver agreement. The reproducibility between observers was considered to be good when the ICC was greater than 0.8000.20 Independent-samples t test was performed to compare the mean ADC value and minimum ADC value in SCC and AC. An ROC analysis was used to compare the diagnostic performance of each criterion for the differentiation of SCC from AC. Thereafter, the area under the curve was evaluated and compared. One-way analysis of variance was used to
DWI for Uterine Cervical Cancer
compare mean ADC value and minimum ADC value among different pathological grade groups. To evaluate the correlation between ADC-based criteria and cellular density, Pearson correlation analysis was performed. Spearman rank correlation coefficient was calculated to evaluate the correlation between ADC-based criteria and pathological grade of tumor, also between cellular density and pathological grade of tumor. A P value of less than 0.05 was considered to indicate a statistically significant difference.
RESULTS Pathological examinations revealed 85 SCC (grade I, 15 cases; grade II, 49 cases; grade III, 21 cases) and 13 AC (grade I, 6 cases; grade II, 5 cases; grade III, 2 cases). Interobserver agreement for mean ADC value (ICC = 0.973) and minimum ADC value (ICC = 0.936) was almost perfect. Uterine cervical cancers were hyperintense to adjacent skeletal muscle on both fat-suppression T2-weighted images and DW images. On pseudocolor ADC maps, they manifested as blue or green regions (Fig. 1; Supplementary Figures 1Y2, available as Supplemental Digital Content at http://links.lww.com/IGC/A295).
Comparison of ADC-Based Criteria Between SCC and AC The mean ADC value and minimum ADC value were (0.868 T 0.121) 10j3 mm2/s, (0.773 T 0.110) 10j3 mm2/s
FIGURE 1. A 46-year-old woman with poorly differentiated SCC of the uterine cervix. A, Axial T2-weighted MR image with fat suppression. The tumor shows high signal intensity with irregular margin. B, The corresponding DWI map. The tumor demonstrates homogeneous high signal intensity. C, Pseudocolor ADC map. The tumor is depicted as an area of blue. The mean ADC value of the tumor is 0.701 10j3 mm2/s, and the minimum ADC value is 0.620 10j3 mm2/s. D, Histological slice reveals poor-differentiated uterine cervical cancer (hematoxylin-eosin stain; original magnification, 200). * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1075
International Journal of Gynecological Cancer
Liu et al
& Volume 25, Number 6, July 2015
Supplementa Digital Content at http://links.lww.com/IGC/A295), and statistical differences were obtained among each group (P G 0.05). Negative correlation was found between mean ADC value and pathological grade of tumor (Spearman rank coefficient, j0.451, P G 0.001), and also between minimum ADC value and pathological grade of tumor (Spearman rank coefficient, j0.501, P G 0.001). Comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.656). There was a significant positive linear correlation between tumor cellular density and the pathological grade of tumor (Spearman rank coefficient, 0.831, P G 0.001). Tumor cellular density correlated negatively with both mean ADC value and minimum ADC value (Pearson coefficient, j0.477, P G 0.001; Pearson coefficient, j0.529, P G 0.001). Comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.631). FIGURE 2. Graphs of ROC curves for the differentiation of SCC from AC. The Az for mean ADC value (0.862) was not significantly greater than that for minimum ADC value (0.861). in SCC; and (1.107 T 0.192) 10j3 mm2/s, (0.974 T 0.172) 10j3 mm2/s in AC. Both mean ADC value and minimum ADC value of SCC were statistically lower than that of AC (P = 0.001; P = 0.000), but there was overlap between them. The Az was 0.862 (standard error, 0.059; 95% confidence interval, 0.747Y0.976) for mean ADC, 0.861 (standard error, 0.057; 95% confidence interval, 0.750Y0.972) for minimum ADC (Fig. 2) in the differentiation between SCC and AC. An ROC analysis revealed that there was no statistically significant difference in the Az values between them (P = 0.990). A mean ADC of less than 0.965 10j3 mm2/s and minimum ADC of less than 0.844 10j3 mm2/s were found to be the most accurate threshold level for distinguishing SCC from AC. When these threshold levels were applied to our study, the sensitivity and specificity for predicting uterine cervical cancer of SCC were 83.5% and 76.9%, and 77.6% and 92.3%, respectively.
Comparison of ADC-Based Criteria and Tumor Cellular Density Among Different Pathological Grade Groups Mean ADC value, minimum ADC value, and tumor cellular density of different pathological grades varied significantly (F = 12.751, P G 0.001; F = 16.408, P G 0.001; F = 114.770, P G 0.001; Table 1, Supplementary Figures 3Y5, available as
DISCUSSION Diffusion-weighted MR imaging enables noninvasive characterization of biological tissues based on the properties of water diffusion. Because of its ability to provide functional and morphological information, there is a growing interest in DWI applications in cancer management. In several malignancies, DWI has been shown to be useful in the discrimination between benign and malignant lesions and in capturing some histopathological features, like tumor pathological subtypes and tumor differentiation.18,21,22 Uterine cervical cancers are usually squamous in origin, but in about 30% of patients are ACs.23 The histological subtype of cervical cancer (SCC, AC, and ASC) is important for making therapeutic decisions and predicting outcomes. The AC/ASC histology is one of the prognostic factors for local recurrence and distant recurrence. For AC/ASC patients, the tumor responses to adjuvant radiotherapy were relatively slow and poor, and this group of patients had a worse survival rate than patients with SCC.24,25 In this study, we found that mean ADC value of SCC was significantly lower than those of AC, in agreement with the result obtained by Kuang et al18 and Liu et al19; also we found that and minimum ADC value SCC was significantly lower than those of AC. The differences in ADC values might reflect differences in histopathological features: cancer nests exhibited increased cellularity and enlarged cell size, which would reduce extracellular spaces; meanwhile, the cancer cells had more organelle, enlarged nuclei, hyperchromatism, and high nuclear-to-cytoplasmic ratio, limiting the diffusion of water molecules in intracellular spaces; because tumor cellularity seemed to be relatively high in SCC, so its ADC values tended to be lower in our study. Therefore, it
TABLE 1. ADC-based criteria and tumor cellular density of different pathological grades Pathologic Grade Grade I Grade II Grade III
1076
No. Patients
Cellular Density
Mean ADC Value
Minimum ADC Value
21 54 23
10.886 T 1.949 15.734 T 2.722 23.220 T 3.358
1.021 T 0.177 0.889 T 0.129 0.813 T 0.121
0.916 T 0.168 0.792 T 0.101 0.710 T 0.107 * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 25, Number 6, July 2015
was feasible to use DWI to distinguish SCC from AC, and this technique would be of great benefit to physicians to choose optimal treatment regimen. Meanwhile, we compared the diagnostic performance for mean ADC value and minimum ADC value, no statistically significant difference was found between them. Therefore, we suggest that ROIs should be placed on the solid parts of the lesions as large as possible to characterize the entire tumor avoiding cystic, necrotic, and hemorrhagic regions which might influence the ADC values. Cellularity, which has been shown to be an important index of tumor grade, is a momentous factor that influences microscopic water diffusion in tumors, because it determines the ratio of extracellular to intracellular space. Cancer, with its increased cellular density, decreased extracellular space, and high nuclear/cytoplasmic ratio, typically shows significant restricted diffusivity compared to surrounding normal tissue, resulting in a low ADC.11 Several DWI studies26Y29 have confirmed that ADC value is related to cell density and inverse correlations between ADC values and tumor cellularity or tumor grade have been noted in several malignancies, including metastatic brain tumors,26 hepatocellular carcinoma,27 lung cancer,28 and breast cancer patients.29 The results from our study showed that mean ADC value of uterine cervical cancer decreased with the increase of the risk grade of the tumors, and significant negative correlation was found between mean ADC value and pathological grade of tumor, in accordance with a previous results19 and recent studies completed by Kuang et al18 and Payne et al.23 Moreover, we also found that minimum ADC value varied significantly among different pathological grades as well, and it was inversely correlated with the pathological grade of tumor. However, we did not observe a significant difference when comparing the correlation coefficients of mean ADC value and minimum ADC value (P = 0.656). In our study, highgrade tumors tended to show higher cellularity; on the contrary, low-grade tumors tended to show lower cellularity. Tumor cellular density of different pathological grades varied significantly and it correlated negatively with both mean ADC value and minimum ADC value, however, comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.631). These phenomena suggest that the differentiation of tumor is likely to be one of the contributing factors affecting ADC value. Therefore, we presume that higher ADC values could be suggestive of lower cellularity and lower grade of the tumor, whereas lower ADC values could suggest higher cellularity and higher grade of the tumor. This study has several limitations. First, only cases of SCC and AC of the uterine cervix were included in this study, however, SCC and AC are the most common subtype of uterine cervical cancer. Second, because not all patients received surgery after MR examination, the biopsy of the tumor for cellularity evaluation might not be the same one for ADC value measurement, and the biopsy was performed before MRI examination in some patients, therefore the matching of the histopathological specimens taken by biopsy to the ADC value measurement of the tumor may have errors. Further investigations with larger number of cases should be performed to verify the results of our preliminary study.
DWI for Uterine Cervical Cancer
CONCLUSIONS Our initial experience suggests that although precise prediction of the histologic type and pathological grade of uterine cervical cancer is not possible because of the overlap among ADC values, ADC measurement may be helpful for the noninvasive and preoperative prediction of histological type and the degree of differentiation of uterine cervical cancer, and we recommend the use of the mean ADC value instead of minimum ADC value to fully reflect the whole tumor.
REFERENCES 1. Chen CC, Lin JC, Jan JS, et al. Definitive intensity-modulated radiation therapy with concurrent chemotherapy for patients with locally advanced cervical cancer. Gynecol Oncol. 2011;122:9Y13. 2. Somoye G, Harry V, Semple S, et al. Early diffusion weighted magnetic resonance imaging can predict survival in women with locally advanced cancer of the cervix treated with combined chemo-radiation. Eur Radiol. 2012;22:2319Y2327. 3. Smith HO, Tiffany MF, Qualls CR, et al. The rising incidence of adenocarcinoma relative to squamous cell carcinoma of the uterine cervix in the United StatesVa 24-year population-based study. Gynecol Oncol. 2000;78:97Y105. 4. Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet. 2001;357:1490Y1493. 5. Sherman ME, Wang SS, Carreon J, et al. Mortality trends for cervical squamous and adenocarcinoma in the United States. Relation to incidence and survival. Cancer. 2005;103:1258Y1264. 6. Bray F, Carstensen B, Møller H, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev. 2005;14:2191Y2199. 7. Irie T, Kigawa J, Minagawa Y, et al. Prognosis and clinicopathological characteristics of Ib-IIb adenocarcinoma of the uterine cervix in patients who have had radical hysterectomy. Eur J Surg Oncol. 2000;26:464Y467. 8. Takeda N, Sakuragi N, Takeda M, et al. Multivariate analysis of histopathologic prognostic factors for invasive cervical cancer treated with radical hysterectomy and systematic retroperitoneal lymphadenectomy. Acta Obstet Gynecol Scand. 2002;81: 1144Y1151. 9. Heatley MK. Systematic review and meta-analysis in anatomic pathology: the value of nuclear DNA content in predicting progression in low grade CIN, the significance of the histological subtype on prognosis in cervical carcinoma. Histol Histopathol. 1999;14:203Y215. 10. Martincich L, Deantoni V, Bertotto I, et al. Correlations between diffusion-weighted imaging and breast cancer biomarkers. Eur Radiol. 2012;22:1519Y1528. 11. Lee NK, Kim S, Seo HI, et al. Diffusion-weighted MR imaging for the differentiation of malignant from benign strictures in the periampullary region. Eur Radiol. 2013;23:1288Y1296. 12. Vandecaveye V, De Keyzer F, Dirix P, et al. Applications of diffusion-weighted magnetic resonance imaging in head and neck squamous cell carcinoma. Neuroradiology. 2010;52:773Y784. 13. Kyriazi S, Nye E, Stamp G, et al. Value of diffusion-weighted imaging for assessing site-specific response of advanced ovarian cancer to neoadjuvant chemotherapy: correlation of apparent diffusion coefficients with epithelial and stromal densities on histology. Cancer Biomark. 2010;7:201Y210.
* 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1077
Liu et al
International Journal of Gynecological Cancer
14. Hara T, Inoue Y, Satoh T, et al. Diffusion-weighted imaging of local recurrent prostate cancer after radiation therapy: comparison with 22-core three-dimensional prostate mapping biopsy. Magn Reson Imaging. 2012;30:1091Y1098. 15. Bonekamp S, Corona-Villalobos CP, Kamel IR. Oncologic applications of diffusion-weighted MRI in the body. J Magn Reson Imaging. 2012;35:257Y279. 16. Chen J, Zhang Y, Liang B, et al. The utility of diffusion-weighted MR imaging in cervical cancer. Eur J Radiol. 2010;74:e101Ye106. 17. Liu Y, Liu H, Bai X, et al. Differentiation of metastatic from non-metastatic lymph nodes in patients with uterine cervical cancer using diffusion-weighted imaging. Gynecol Oncol. 2011;122:19Y24. 18. Kuang F, Ren J, Zhong Q, et al. The value of apparent diffusion coefficient in the assessment of cervical cancer. Eur Radiol. 2013;23:1050Y1058. 19. Liu Y, Bai R, Sun H, et al. Diffusion-weighted magnetic resonance imaging of uterine cervical cancer. J Comput Assist Tomogr. 2009;33:858Y862. 20. Park SO, Kim JK, Kim KA, et al. Relative apparent diffusion coefficient: determination of reference site and validation of benefit for detecting metastatic lymph nodes in uterine cervical cancer. J Magn Reson Imaging. 2009;29:383Y390. 21. Yu X, Lin M, Ouyang H, et al. Application of ADC measurement in characterization of renal cell carcinomas with different pathological types and grades by 3.0 T diffusion-weighted MRI. Eur J Radiol. 2012;81:3061Y3066. 22. Watanabe Y, Yamasaki F, Kajiwara Y, et al. Preoperative histological grading of meningiomas using apparent diffusion coefficient at 3 T MRI. Eur J Radiol. 2013;82:658Y663.
1078
& Volume 25, Number 6, July 2015
23. Payne GS, Schmidt M, Morgan VA, et al. Evaluation of magnetic resonance diffusion and spectroscopy measurements as predictive biomarkers in stage 1 cervical cancer. Gynecol Oncol. 2010;116:246Y252. 24. Huang YT, Wang CC, Tsai CS, et al. Clinical behaviors and outcomes for adenocarcinoma or adenosquamous carcinoma of cervix treated by radical hysterectomy and adjuvant radiotherapy or chemoradiotherapy. Int J Radiat Oncol Biol Phys. 2012;84:420Y427. 25. Hong JH, Tsai CS, Wang CC, et al. Comparison of clinical behaviors and responses to radiation between squamous cell carcinomas and adenocarcinomas/adenosquamous carcinomas of the cervix. Chang Gung Med J. 2000;23:396Y404. 26. Hayashida Y, Hirai T, Morishita S, et al. Diffusion-weighted imaging of metastatic brain tumors: comparison with histologic type and tumor cellularity. AJNR Am J Neuroradiol. 2006;27:1419Y1425. 27. Heo SH, Jeong YY, Shin SS, et al. Apparent diffusion coefficient value of diffusion-weighted imaging for hepatocellular carcinoma: correlation with the histologic differentiation and the expression of vascular endothelial growth factor. Korean J Radiol. 2010;11:295Y303. 28. Matoba M, Tonami H, Kondou T, et al. Lung carcinoma: diffusion-weighted mr imagingVpreliminary evaluation with apparent diffusion coefficient. Radiology. 2007;243:570Y577. 29. Costantini M, Belli P, Rinaldi P, et al. Diffusion-weighted imaging in breast cancer: relationship between apparent diffusion coefficient and tumour aggressiveness. Clin Radiol. 2010;65:1005Y1012.
* 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Clinical Application of Diffusion-Weighted Magnetic Resonance Imaging in Uterine Cervical Cancer Ying Liu, MD,* Zhaoxiang Ye, MD,* Haoran Sun, MD,Þ and Renju Bai, MDÞ
Objective: This study aimed to investigate the application value of apparent diffusion coefficient (ADC) values in evaluating histological type as well as pathologic grade of uterine cervical cancer; and to investigate whether ADC values could reflect tumor cellular density. Methods: Ninety-eight patients with histopathologically proven uterine cervical cancer were included in this study. Mean ADC value and minimum ADC value of the tumor were measured. Tumor cellular density was counted using colored multifunction imaging analyzing system. Results: Both mean ADC value and minimum ADC value of squamous cell carcinoma were significantly lower than that of adenocarcinoma (P = 0.001; P = 0.000). Using mean ADC criteria (e0.965 10j3 mm2/s) and minimum ADC criteria (e0.844 10j3 mm2/s), the sensitivity and specificity for differentiating squamous cell carcinoma from adenocarcinoma were 83.5% and 76.9%, and 77.6% and 92.3%, respectively. Receiver operating characteristic analysis revealed that there was no statistically significant difference in the Az values between them (P = 0.990). Tumor cellular density, mean ADC value, and minimum ADC value of different pathological grade varied significantly (P = 0.000, P = 0.000, P = 0.000). There was a significant positive linear correlation between tumor cellular density and pathological grade of tumor (P = 0.000). Both mean ADC value and minimum ADC value correlated negatively with cellular density (P = 0.000, P = 0.000) and the pathological grade of tumor (P = 0.000, P = 0.000). Comparisons of correlation coefficients showed no significant differences (P = 0.656, P = 0.631). Conclusions: Diffusion-weighted magnetic resonance imaging has a potential ability to indicate the histologic type of uterine cervical cancer. Apparent diffusion coefficient measurements of uterine cervical cancer can represent tumor cellular density, thus providing a new method for evaluating the pathological grade of tumor. Key Words: Uterine cervical cancer, Diffusion-weighted MR imaging, Cellularity density, Pathological grade Received August 13, 2014, and in revised form March 2, 2015. Accepted for publication March 19, 2015. (Int J Gynecol Cancer 2015;25: 1073Y1078)
*Department of Radiology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center of Cancer, Key Laboratory of Cancer Prevention and Therapy; and †Department of Radiology, General Hospital of Tianjin Medical University, Tianjin, China. Copyright * 2015 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000472 International Journal of Gynecological Cancer
Address correspondence and reprint requests to Ying Liu, MD, Department of Radiology, Tianjin Medical University Cancer Institute and Hospital, Huan-Hu-Xi Rd, Ti-Yuan-Bei, He Xi District, Tianjin 300060, China. E-mail: [email protected]. The authors declare no conflicts of interest. Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net).
& Volume 25, Number 6, July 2015
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1073
International Journal of Gynecological Cancer
Liu et al
cervical cancer is one of the most common gyneU terine cological tumors diagnosed in women worldwide, and is 1
the second most common cause of cancer-related deaths in women living in developing countries.2 Despite major advances in screening and prevention, the relative proportion of cervical adenocarcinoma (AC) and adenosquamous carcinoma (ASC), as compared with squamous cell carcinoma (SCC), has increased.3Y6 Some studies have reported that AC and poorly differentiated tumors show increased likelihood of metastasis and poorer outcomes than their squamous and well-differentiated counterparts, respectively.7Y9 Diffusion-weighted magnetic resonance (MR) imaging (DWI) is a noninvasive technique that supplies information of water proton mobility. It provides information on extracellular space tortuosity, tissue cellularity, and cellular membrane integrity, which enables the characterization of tissue at microscopic level and makes subtle abnormalities more obvious.10Y12 It is being incorporated into oncological imaging practice as a possible tool for characterizing tissue (ie, discriminating between benign and malignant lesions, distinguishing among tumor types and assessing tumor grade, monitoring response to therapy, and detecting recurrent cancer).13Y16 To date, DWI has become a routine imaging protocol for the female pelvis, because it is so quick to perform, and can be simply adopted on most current MRI scanners without any additional new equipment and intravenous contrast agents. Clinical applications of DWI for uterine cervical cancer predominantly focused on tumor detection, staging of disease, differentiating between metastatic and nonYmetastatic lymph nodes, and treatment evaluation.17,18 However, little information is available for capturing some histopathological features of uterine cervical cancer by using DWI. In previous studies,18,19 it has been confirmed that mean apparent diffusion coefficient (ADC) value of uterine cervical cancer could indicate the histologic type and represent tumor cellular density to some extent, however, they did not compare the diagnostic performance between mean ADC value and minimum ADC value. Thus, the purpose of this study was to evaluate whether ADC value could reflect the histological type and tumor differentiation with large samples, meanwhile, investigate the relationship between ADC value and tumor cellular density.
MATERIALS AND METHODS Patient Population Our study received institutional ethics committee approval and informed consent was obtained from all patients. Inclusion criteria consisted of (1) International Federation of Gynecology and Obstetrics stage based on clinical examination ranges from II to IV, (2) no previous radiation or concurrent chemoradiation therapy for uterine cervical cancer before MRI examination. Exclusion criteria consisted of (1) unable to get histologically (biopsy or surgery) proven of uterine cervical cancer, (2) receiving antineoplastic therapy after MRI examination and then undergo surgical operation or biopsy. Ninety-eight female patients (mean age, 49 years; age range, 25Y80 years) with uterine cervical cancer enrolled in this study. Uterine cervical cancer was confirmed in all 98 patients, either by surgery (37 cases) or by biopsy (61 cases).
1074
& Volume 25, Number 6, July 2015
MR Examination Magnetic resonance examinations were performed before surgery and the initiation of concurrent chemoradiation. To limit bowel motion, patients were recommended to fast for 4 to 6 hours before imaging. All MR examinations were performed using a 1.5-T unit (Twin Excite, GE Healthcare, USA) with Torso phased-array body coil. Before DWI, conventional T2-weighted fast spin-echo in the sagittal and transverse planes (TR/TE, 4000 ms/85 ms; matrix size, 320 224; band width, 31.25 Hz/pixel; field of view, 36 cm; number of excitation, 2; slice thickness, 6 mm; gap, 1 mm), T2-weighted fast spin-echo with fat suppression in the transverse plane (the parameters were the same as T2-weighted image) and T1-weighted spin-echo in transverse plane (TR/TE, 500 ms/ 20 ms; matrix size, 320 160; band width, 31.25 Hz/pixel; field of view, 36 cm; number of excitation, 2; slice thickness, 6 mm; gap, 1 mm) were obtained. Transverse diffusion-weighted images were obtained using a nonYbreath-hold single-shot spin-echo echo-planar imaging sequence and array spatial sensitivity encoding technique with the following parameters: repetition time of 4000 milliseconds; echo time of 58.5 milliseconds; matrix size, 128 128; field of view, 36 cm; number of excitation, 4; slice thickness, 6 mm; gap, 1 mm; R factor, 2; phase-encoding direction, anteroposterior; and b value of 0 and 1000 s/mm2. The diffusion-weighting gradients were applied in all 3 orthogonal directions. The scanning time of DWI was 1 minute and 4 seconds.
MR Image Analysis Magnetic resonance images were analyzed in consensus by 2 radiologists. Both observers were blinded to each other’s MR assessments and clinical information including histological type and pathological grade of tumor. Mean ADC value and minimum ADC value were extracted from the manual placement of the region of interest (ROI) on the ADC map. The central slice which could manifest the largest part of tumor was selected. Mean ADC was obtained by a single measurement covering the solid portion of tumor as large as possible, avoiding volume averaging with cystic or degenerative regions that might influence the quantitative data; whereas a minimum ADC was extracted from the manual placement of 5 circular ROIs in the former ROI, and the ROI with the lowest ADC value was selected as the minimum ADC value.
Histopathological Evaluation Ninety-eight cases of uterine cervical cancer were analyzed for histopathological evaluation. A single experienced pathologist with 15 years of experience, who was blinded to MR results, coded and evaluated all slides. Formalin-fixed paraffin-embedded tissue blocks of cervical biopsy specimens taken pretreatment or of resection specimens taken after surgery, were sectioned at a thickness of 4 Km, and then processed for conventional histological assessment by hematoxylin and eosin staining to verify the morphological diagnosis of uterine cervical cancer and to evaluate its differentiation. The differentiation of uterine cervical cancer was classified into well (grade I), moderate (grade II), and poor * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 25, Number 6, July 2015
(grade III). When different tumor grades coexisted within a tumor, the more predominant differentiation of the tumor was selected. For each specimen (original magnification, 200), 5 visual fields that showed representative pathological findings in each carcinoma were chosen by 1 pathologist. Cellular density of each specimen was measured in colored multifunction imaging analyzing system. Tumor cellular density was defined as ratios of the total area of tumor cell nuclei divided by the area of sample image in each specimen, expressed by percentage.
Statistical Analysis Statistical analyses were performed by using statistical software (SPSS, version 13.0) with the exception of the comparison of areas under independent ROC curves and correlation coefficients, and the assessment of interobserver agreement, which were performed using MedCalc for windows (version 11.3.8.0; MedCalc software). The intraclass correlation coefficient (ICC) was used to assess the interobserver agreement. The reproducibility between observers was considered to be good when the ICC was greater than 0.8000.20 Independent-samples t test was performed to compare the mean ADC value and minimum ADC value in SCC and AC. An ROC analysis was used to compare the diagnostic performance of each criterion for the differentiation of SCC from AC. Thereafter, the area under the curve was evaluated and compared. One-way analysis of variance was used to
DWI for Uterine Cervical Cancer
compare mean ADC value and minimum ADC value among different pathological grade groups. To evaluate the correlation between ADC-based criteria and cellular density, Pearson correlation analysis was performed. Spearman rank correlation coefficient was calculated to evaluate the correlation between ADC-based criteria and pathological grade of tumor, also between cellular density and pathological grade of tumor. A P value of less than 0.05 was considered to indicate a statistically significant difference.
RESULTS Pathological examinations revealed 85 SCC (grade I, 15 cases; grade II, 49 cases; grade III, 21 cases) and 13 AC (grade I, 6 cases; grade II, 5 cases; grade III, 2 cases). Interobserver agreement for mean ADC value (ICC = 0.973) and minimum ADC value (ICC = 0.936) was almost perfect. Uterine cervical cancers were hyperintense to adjacent skeletal muscle on both fat-suppression T2-weighted images and DW images. On pseudocolor ADC maps, they manifested as blue or green regions (Fig. 1; Supplementary Figures 1Y2, available as Supplemental Digital Content at http://links.lww.com/IGC/A295).
Comparison of ADC-Based Criteria Between SCC and AC The mean ADC value and minimum ADC value were (0.868 T 0.121) 10j3 mm2/s, (0.773 T 0.110) 10j3 mm2/s
FIGURE 1. A 46-year-old woman with poorly differentiated SCC of the uterine cervix. A, Axial T2-weighted MR image with fat suppression. The tumor shows high signal intensity with irregular margin. B, The corresponding DWI map. The tumor demonstrates homogeneous high signal intensity. C, Pseudocolor ADC map. The tumor is depicted as an area of blue. The mean ADC value of the tumor is 0.701 10j3 mm2/s, and the minimum ADC value is 0.620 10j3 mm2/s. D, Histological slice reveals poor-differentiated uterine cervical cancer (hematoxylin-eosin stain; original magnification, 200). * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1075
International Journal of Gynecological Cancer
Liu et al
& Volume 25, Number 6, July 2015
Supplementa Digital Content at http://links.lww.com/IGC/A295), and statistical differences were obtained among each group (P G 0.05). Negative correlation was found between mean ADC value and pathological grade of tumor (Spearman rank coefficient, j0.451, P G 0.001), and also between minimum ADC value and pathological grade of tumor (Spearman rank coefficient, j0.501, P G 0.001). Comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.656). There was a significant positive linear correlation between tumor cellular density and the pathological grade of tumor (Spearman rank coefficient, 0.831, P G 0.001). Tumor cellular density correlated negatively with both mean ADC value and minimum ADC value (Pearson coefficient, j0.477, P G 0.001; Pearson coefficient, j0.529, P G 0.001). Comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.631). FIGURE 2. Graphs of ROC curves for the differentiation of SCC from AC. The Az for mean ADC value (0.862) was not significantly greater than that for minimum ADC value (0.861). in SCC; and (1.107 T 0.192) 10j3 mm2/s, (0.974 T 0.172) 10j3 mm2/s in AC. Both mean ADC value and minimum ADC value of SCC were statistically lower than that of AC (P = 0.001; P = 0.000), but there was overlap between them. The Az was 0.862 (standard error, 0.059; 95% confidence interval, 0.747Y0.976) for mean ADC, 0.861 (standard error, 0.057; 95% confidence interval, 0.750Y0.972) for minimum ADC (Fig. 2) in the differentiation between SCC and AC. An ROC analysis revealed that there was no statistically significant difference in the Az values between them (P = 0.990). A mean ADC of less than 0.965 10j3 mm2/s and minimum ADC of less than 0.844 10j3 mm2/s were found to be the most accurate threshold level for distinguishing SCC from AC. When these threshold levels were applied to our study, the sensitivity and specificity for predicting uterine cervical cancer of SCC were 83.5% and 76.9%, and 77.6% and 92.3%, respectively.
Comparison of ADC-Based Criteria and Tumor Cellular Density Among Different Pathological Grade Groups Mean ADC value, minimum ADC value, and tumor cellular density of different pathological grades varied significantly (F = 12.751, P G 0.001; F = 16.408, P G 0.001; F = 114.770, P G 0.001; Table 1, Supplementary Figures 3Y5, available as
DISCUSSION Diffusion-weighted MR imaging enables noninvasive characterization of biological tissues based on the properties of water diffusion. Because of its ability to provide functional and morphological information, there is a growing interest in DWI applications in cancer management. In several malignancies, DWI has been shown to be useful in the discrimination between benign and malignant lesions and in capturing some histopathological features, like tumor pathological subtypes and tumor differentiation.18,21,22 Uterine cervical cancers are usually squamous in origin, but in about 30% of patients are ACs.23 The histological subtype of cervical cancer (SCC, AC, and ASC) is important for making therapeutic decisions and predicting outcomes. The AC/ASC histology is one of the prognostic factors for local recurrence and distant recurrence. For AC/ASC patients, the tumor responses to adjuvant radiotherapy were relatively slow and poor, and this group of patients had a worse survival rate than patients with SCC.24,25 In this study, we found that mean ADC value of SCC was significantly lower than those of AC, in agreement with the result obtained by Kuang et al18 and Liu et al19; also we found that and minimum ADC value SCC was significantly lower than those of AC. The differences in ADC values might reflect differences in histopathological features: cancer nests exhibited increased cellularity and enlarged cell size, which would reduce extracellular spaces; meanwhile, the cancer cells had more organelle, enlarged nuclei, hyperchromatism, and high nuclear-to-cytoplasmic ratio, limiting the diffusion of water molecules in intracellular spaces; because tumor cellularity seemed to be relatively high in SCC, so its ADC values tended to be lower in our study. Therefore, it
TABLE 1. ADC-based criteria and tumor cellular density of different pathological grades Pathologic Grade Grade I Grade II Grade III
1076
No. Patients
Cellular Density
Mean ADC Value
Minimum ADC Value
21 54 23
10.886 T 1.949 15.734 T 2.722 23.220 T 3.358
1.021 T 0.177 0.889 T 0.129 0.813 T 0.121
0.916 T 0.168 0.792 T 0.101 0.710 T 0.107 * 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 25, Number 6, July 2015
was feasible to use DWI to distinguish SCC from AC, and this technique would be of great benefit to physicians to choose optimal treatment regimen. Meanwhile, we compared the diagnostic performance for mean ADC value and minimum ADC value, no statistically significant difference was found between them. Therefore, we suggest that ROIs should be placed on the solid parts of the lesions as large as possible to characterize the entire tumor avoiding cystic, necrotic, and hemorrhagic regions which might influence the ADC values. Cellularity, which has been shown to be an important index of tumor grade, is a momentous factor that influences microscopic water diffusion in tumors, because it determines the ratio of extracellular to intracellular space. Cancer, with its increased cellular density, decreased extracellular space, and high nuclear/cytoplasmic ratio, typically shows significant restricted diffusivity compared to surrounding normal tissue, resulting in a low ADC.11 Several DWI studies26Y29 have confirmed that ADC value is related to cell density and inverse correlations between ADC values and tumor cellularity or tumor grade have been noted in several malignancies, including metastatic brain tumors,26 hepatocellular carcinoma,27 lung cancer,28 and breast cancer patients.29 The results from our study showed that mean ADC value of uterine cervical cancer decreased with the increase of the risk grade of the tumors, and significant negative correlation was found between mean ADC value and pathological grade of tumor, in accordance with a previous results19 and recent studies completed by Kuang et al18 and Payne et al.23 Moreover, we also found that minimum ADC value varied significantly among different pathological grades as well, and it was inversely correlated with the pathological grade of tumor. However, we did not observe a significant difference when comparing the correlation coefficients of mean ADC value and minimum ADC value (P = 0.656). In our study, highgrade tumors tended to show higher cellularity; on the contrary, low-grade tumors tended to show lower cellularity. Tumor cellular density of different pathological grades varied significantly and it correlated negatively with both mean ADC value and minimum ADC value, however, comparison of correlation coefficients showed that there was no statistical difference between them (P = 0.631). These phenomena suggest that the differentiation of tumor is likely to be one of the contributing factors affecting ADC value. Therefore, we presume that higher ADC values could be suggestive of lower cellularity and lower grade of the tumor, whereas lower ADC values could suggest higher cellularity and higher grade of the tumor. This study has several limitations. First, only cases of SCC and AC of the uterine cervix were included in this study, however, SCC and AC are the most common subtype of uterine cervical cancer. Second, because not all patients received surgery after MR examination, the biopsy of the tumor for cellularity evaluation might not be the same one for ADC value measurement, and the biopsy was performed before MRI examination in some patients, therefore the matching of the histopathological specimens taken by biopsy to the ADC value measurement of the tumor may have errors. Further investigations with larger number of cases should be performed to verify the results of our preliminary study.
DWI for Uterine Cervical Cancer
CONCLUSIONS Our initial experience suggests that although precise prediction of the histologic type and pathological grade of uterine cervical cancer is not possible because of the overlap among ADC values, ADC measurement may be helpful for the noninvasive and preoperative prediction of histological type and the degree of differentiation of uterine cervical cancer, and we recommend the use of the mean ADC value instead of minimum ADC value to fully reflect the whole tumor.
REFERENCES 1. Chen CC, Lin JC, Jan JS, et al. Definitive intensity-modulated radiation therapy with concurrent chemotherapy for patients with locally advanced cervical cancer. Gynecol Oncol. 2011;122:9Y13. 2. Somoye G, Harry V, Semple S, et al. Early diffusion weighted magnetic resonance imaging can predict survival in women with locally advanced cancer of the cervix treated with combined chemo-radiation. Eur Radiol. 2012;22:2319Y2327. 3. Smith HO, Tiffany MF, Qualls CR, et al. The rising incidence of adenocarcinoma relative to squamous cell carcinoma of the uterine cervix in the United StatesVa 24-year population-based study. Gynecol Oncol. 2000;78:97Y105. 4. Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet. 2001;357:1490Y1493. 5. Sherman ME, Wang SS, Carreon J, et al. Mortality trends for cervical squamous and adenocarcinoma in the United States. Relation to incidence and survival. Cancer. 2005;103:1258Y1264. 6. Bray F, Carstensen B, Møller H, et al. Incidence trends of adenocarcinoma of the cervix in 13 European countries. Cancer Epidemiol Biomarkers Prev. 2005;14:2191Y2199. 7. Irie T, Kigawa J, Minagawa Y, et al. Prognosis and clinicopathological characteristics of Ib-IIb adenocarcinoma of the uterine cervix in patients who have had radical hysterectomy. Eur J Surg Oncol. 2000;26:464Y467. 8. Takeda N, Sakuragi N, Takeda M, et al. Multivariate analysis of histopathologic prognostic factors for invasive cervical cancer treated with radical hysterectomy and systematic retroperitoneal lymphadenectomy. Acta Obstet Gynecol Scand. 2002;81: 1144Y1151. 9. Heatley MK. Systematic review and meta-analysis in anatomic pathology: the value of nuclear DNA content in predicting progression in low grade CIN, the significance of the histological subtype on prognosis in cervical carcinoma. Histol Histopathol. 1999;14:203Y215. 10. Martincich L, Deantoni V, Bertotto I, et al. Correlations between diffusion-weighted imaging and breast cancer biomarkers. Eur Radiol. 2012;22:1519Y1528. 11. Lee NK, Kim S, Seo HI, et al. Diffusion-weighted MR imaging for the differentiation of malignant from benign strictures in the periampullary region. Eur Radiol. 2013;23:1288Y1296. 12. Vandecaveye V, De Keyzer F, Dirix P, et al. Applications of diffusion-weighted magnetic resonance imaging in head and neck squamous cell carcinoma. Neuroradiology. 2010;52:773Y784. 13. Kyriazi S, Nye E, Stamp G, et al. Value of diffusion-weighted imaging for assessing site-specific response of advanced ovarian cancer to neoadjuvant chemotherapy: correlation of apparent diffusion coefficients with epithelial and stromal densities on histology. Cancer Biomark. 2010;7:201Y210.
* 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1077
Liu et al
International Journal of Gynecological Cancer
14. Hara T, Inoue Y, Satoh T, et al. Diffusion-weighted imaging of local recurrent prostate cancer after radiation therapy: comparison with 22-core three-dimensional prostate mapping biopsy. Magn Reson Imaging. 2012;30:1091Y1098. 15. Bonekamp S, Corona-Villalobos CP, Kamel IR. Oncologic applications of diffusion-weighted MRI in the body. J Magn Reson Imaging. 2012;35:257Y279. 16. Chen J, Zhang Y, Liang B, et al. The utility of diffusion-weighted MR imaging in cervical cancer. Eur J Radiol. 2010;74:e101Ye106. 17. Liu Y, Liu H, Bai X, et al. Differentiation of metastatic from non-metastatic lymph nodes in patients with uterine cervical cancer using diffusion-weighted imaging. Gynecol Oncol. 2011;122:19Y24. 18. Kuang F, Ren J, Zhong Q, et al. The value of apparent diffusion coefficient in the assessment of cervical cancer. Eur Radiol. 2013;23:1050Y1058. 19. Liu Y, Bai R, Sun H, et al. Diffusion-weighted magnetic resonance imaging of uterine cervical cancer. J Comput Assist Tomogr. 2009;33:858Y862. 20. Park SO, Kim JK, Kim KA, et al. Relative apparent diffusion coefficient: determination of reference site and validation of benefit for detecting metastatic lymph nodes in uterine cervical cancer. J Magn Reson Imaging. 2009;29:383Y390. 21. Yu X, Lin M, Ouyang H, et al. Application of ADC measurement in characterization of renal cell carcinomas with different pathological types and grades by 3.0 T diffusion-weighted MRI. Eur J Radiol. 2012;81:3061Y3066. 22. Watanabe Y, Yamasaki F, Kajiwara Y, et al. Preoperative histological grading of meningiomas using apparent diffusion coefficient at 3 T MRI. Eur J Radiol. 2013;82:658Y663.
1078
& Volume 25, Number 6, July 2015
23. Payne GS, Schmidt M, Morgan VA, et al. Evaluation of magnetic resonance diffusion and spectroscopy measurements as predictive biomarkers in stage 1 cervical cancer. Gynecol Oncol. 2010;116:246Y252. 24. Huang YT, Wang CC, Tsai CS, et al. Clinical behaviors and outcomes for adenocarcinoma or adenosquamous carcinoma of cervix treated by radical hysterectomy and adjuvant radiotherapy or chemoradiotherapy. Int J Radiat Oncol Biol Phys. 2012;84:420Y427. 25. Hong JH, Tsai CS, Wang CC, et al. Comparison of clinical behaviors and responses to radiation between squamous cell carcinomas and adenocarcinomas/adenosquamous carcinomas of the cervix. Chang Gung Med J. 2000;23:396Y404. 26. Hayashida Y, Hirai T, Morishita S, et al. Diffusion-weighted imaging of metastatic brain tumors: comparison with histologic type and tumor cellularity. AJNR Am J Neuroradiol. 2006;27:1419Y1425. 27. Heo SH, Jeong YY, Shin SS, et al. Apparent diffusion coefficient value of diffusion-weighted imaging for hepatocellular carcinoma: correlation with the histologic differentiation and the expression of vascular endothelial growth factor. Korean J Radiol. 2010;11:295Y303. 28. Matoba M, Tonami H, Kondou T, et al. Lung carcinoma: diffusion-weighted mr imagingVpreliminary evaluation with apparent diffusion coefficient. Radiology. 2007;243:570Y577. 29. Costantini M, Belli P, Rinaldi P, et al. Diffusion-weighted imaging in breast cancer: relationship between apparent diffusion coefficient and tumour aggressiveness. Clin Radiol. 2010;65:1005Y1012.
* 2015 IGCS and ESGO
Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
