CME JOURNAL OF MAGNETIC RESONANCE IMAGING 00:00–00 (2013)
Original Research
Correlation of Apparent Diffusion Coefficient Value With Prognostic Parameters of Endometrioid Carcinoma Chie Inoue, MD,1* Shinya Fujii, MD, PhD,1 Sachi Kaneda, MD, PhD,1 Takeru Fukunaga, MD,1 Toshio Kaminou, MD, PhD,1 Junzo Kigawa, MD, PhD,2 Tasuku Harada, MD, PhD,2 and Toshihide Ogawa, MD, PhD1 ENDOMETRIAL CARCINOMA is one of the most common gynecological malignancies (1). Several factors are related to prognosis, such as International Federation of Gynecology and Obstetrics (FIGO) stage (including depth of myometrial invasion and presence of cervical involvement), histological subtype and grade, lymphovascular invasion, and presence of lymph node metastases (2–5). Although the FIGO staging system of endometrial carcinoma is a surgicopathologic staging system, appropriate preoperative assessment of the invasiveness of endometrial carcinoma is important for surgical planning. Magnetic resonance imaging (MRI) is currently the best examination method to evaluate the local invasiveness of uterine cancers preoperatively (6–9). Diffusion-weighted imaging (DWI) is a functional imaging modality that depicts the microscopic mobility of water. DWI can also provide the apparent diffusion coefficient (ADC) for a tissue, representing a quantitative index describing microscopic water diffusivity (10). In general, malignant tumors show a lower ADC, facilitating the differentiation of malignant and benign lesions (11–20). This is probably related to a combination of higher cellularity, tissue disorganization, and increased extracellular space tortuosity, all contributing to reduced motion of water (21). Correlations with cellularity have been found for some primary and secondary neoplasms (11,22–26). On the other hand, even when using ADC there is also the difficulty of differential diagnosis of malignant and benign lesion, such as lymph node metastasis and ovarian lesion (27,28). The usefulness of measuring ADC in patients with endometrial carcinoma has been reported in several studies (4,29–31); ADC is useful in differentiating between normal and cancerous tissue of the endometrium and cervix, and in differentiating malignant from benign uterine endometrial cavity lesions. Several authors have also reported correlations between ADC and prognostic parameters of endometrioid carcinoma (3–5,32–35), although the results were controversial.
Purpose: To correlate the apparent diffusion coefficient (ADC) of endometrioid carcinoma with histological tumor grade and degree of myometrial invasion. Materials and Methods: 3T diffusion-weighted (DW) magnetic resonance (MR) images of 63 patients were retrospectively reviewed. Two readers measured tumor ADC according to a freehand region of interest (ROI) and a round ROI. Mean and minimum ADCs were correlated with prognostic parameters. Results: The minimum ADC was 0.64 103mm2/s for grade 1 (G1, n ¼ 42), 0.62 for grade 2 (G2, n ¼ 14), 0.46 for grade 3 (G3, n ¼ 7) on freehand ROI. There were significant differences between G1 and G3 (P ¼ 0.007), and G2 and G3 (P ¼ 0.038). No significant correlation was found between tumor grade and mean ADC (0.85 for G1, 0.82 for G2, and 0.72 for G3, P ¼ 0.166). The minimum ADC was significantly lower for patients with deep (n ¼ 21, 0.54) than for those with superficial (n ¼ 39, 0.65) myometrial invasion. Conversely, mean ADC did not differ significantly (0.84 for superficial and 0.78 for deep myometrial invasion, P ¼ 0.081). The same tendency was shown on round ROI. Conclusion: The minimum ADC correlates with prognostic parameters of endometrial carcinoma more strongly than mean ADC. Lower minimum ADC is associated with higher histological tumor grade and higher degree of myometrial invasion. Key Words: MR; diffusion; apparent diffusion coefficient; ADC; endometrioid carcinoma; prognostic parameters J. Magn. Reson. Imaging 2013;00:000–000. C 2013 Wiley Periodicals, Inc. V
1 Division of Radiology, Department of Pathophysiological and Therapeutic Science, Faculty of Medicine, Tottori University, Yonago, Japan. 2 Division of Reproductive-Perinatal Medicine and Gynecological Oncology, Department of Surgery, Faculty of Medicine, Tottori University, Yonago, Japan. *Address reprint requests to: C.I., Division of Radiology, Department of Pathophysiological and Therapeutic Science, Faculty of Medicine, Tottori University, 36-1, Nishi-cho, Yonago, Tottori 683–8504, Japan. E-mail: [email protected] Received August 26, 2013; Accepted November 14, 2013. DOI 10.1002/jmri.24534 View this article online at wileyonlinelibrary.com. C 2013 Wiley Periodicals, Inc. V
1
2
Inoue et al.
Figure 1. ADC measurements of the tumor were performed on an ADC map according to freehand ROI (a) and round ROI (b). Tumor was defined as a focal mass showing high signal intensity on axial DW image (c) compared with the signal of adjacent tissue. The ROIs were carefully set not to involve necrosis by referring to T2-weighted image (d) and contrast enhanced T1-weighted image (e).
The purpose of this study was to examine the correlations between ADC and prognostic parameters of endometrioid carcinoma such as histological grade, degree of myometrial invasion, and the presence of lymph node metastases. MATERIALS AND METHODS Patient Population An initial total of 73 patients were studied retrospectively. All patients had undergone MR examinations including DWI using a 3.0-T MR system between August 2006 and December 2011 at our institution and had received pathological confirmation of endometrial carcinoma. Ten of these patients were excluded because the tumors were too small to be clearly demonstrated on DWI. As a result, a final total of 63 patients (age range, 32–92 years; mean, 60 years) were enrolled. Of the 63 endometrioid adenocarcinomas, 42 were well-differentiated (grade 1), 14 were moderately differentiated (grade 2), and seven were poorly differentiated (grade 3). Three patients did not receive surgical therapy because of distant metastases and pathological confirmation was obtained by biopsy. Pelvic lymph node resection was not performed on seven patients because of a clinical state such as advanced age or preexisting disease. The Institutional Review Board approved the study protocols and waived the requirement for informed consent.
weighted spoiled gradient recalled (SPGR) images (repetition time [TR], 250 msec; echo time [TE], 2.1 msec; flip angle [FA], 75 ; acquisition time, 1 min 30 sec), axial and sagittal T2-weighted fast spin echo (FSE) images (TR, 6500 msec; TE, 100 msec; section thickness, 5 mm; intersection gap, 1.5 mm; acquisition time, 3–3.5 min), and axial and/or sagittal SPGR images with fat suppression (TR, 320 msec; TE, 2.1 msec; FA, 75 ; acquisition time, 1 min 50 sec). Axial DW images were then obtained (TR, 5500–6000 msec; TE, 60–62 msec; inversion time, 200 msec; b factors, 0 and 1000 s/mm2; 112 128 matrix; field of view, 400 mm; section thickness, 4–6 mm with no gap; SENSE reduction factor, 2; signals acquired, 4; acquisition time, 3.5 min). STIR was used for fat suppression in a single spin echo planar imaging sequence, with free breathing during acquisition. Motion-probing gradient pulses were placed in the three orthogonal planes. Isotropic DWI was generated using three orthogonal-axis images. We then obtained axial FSE images as reference images. Parameters for FSE with parallel imaging were as follows: TR, 4500–5000 msec; TE, 100 msec; FA, 160 ; echo spacing, 7.4 msec; echo train length, 20; 512 256 matrix; section thickness, 4–6 mm with no gap; number of excitations, 0.5; free breathing; acquisition time, 50–60 sec. Before imaging, all patients received intramuscular administration of 20 mg of butyl-scopolamine (Buscopan; Nippon Boehringer Ingelheim, Tokyo, Japan) to prevent peristalsis artifacts, except when contraindicated.
MR Technique MRI was performed using a 3.0-T MR system (Signa EXCITE HD; GE Medical Systems, Milwaukee, WI) with phased array coils. Routine pelvic MR images were acquired as follows: axial and sagittal T1-
Image Analysis MR images were independently analyzed by two radiologists (C.I. and S.K.; 8 and 6 years of experience in
Correlation of ADC With Prognostic Parameters
3
Table 1 Mean ADC of Endometrioid Carcinoma With Various Prognostic Factors n
Freehand ROI
Round ROI
G1 G2 G3
Original Research
Correlation of Apparent Diffusion Coefficient Value With Prognostic Parameters of Endometrioid Carcinoma Chie Inoue, MD,1* Shinya Fujii, MD, PhD,1 Sachi Kaneda, MD, PhD,1 Takeru Fukunaga, MD,1 Toshio Kaminou, MD, PhD,1 Junzo Kigawa, MD, PhD,2 Tasuku Harada, MD, PhD,2 and Toshihide Ogawa, MD, PhD1 ENDOMETRIAL CARCINOMA is one of the most common gynecological malignancies (1). Several factors are related to prognosis, such as International Federation of Gynecology and Obstetrics (FIGO) stage (including depth of myometrial invasion and presence of cervical involvement), histological subtype and grade, lymphovascular invasion, and presence of lymph node metastases (2–5). Although the FIGO staging system of endometrial carcinoma is a surgicopathologic staging system, appropriate preoperative assessment of the invasiveness of endometrial carcinoma is important for surgical planning. Magnetic resonance imaging (MRI) is currently the best examination method to evaluate the local invasiveness of uterine cancers preoperatively (6–9). Diffusion-weighted imaging (DWI) is a functional imaging modality that depicts the microscopic mobility of water. DWI can also provide the apparent diffusion coefficient (ADC) for a tissue, representing a quantitative index describing microscopic water diffusivity (10). In general, malignant tumors show a lower ADC, facilitating the differentiation of malignant and benign lesions (11–20). This is probably related to a combination of higher cellularity, tissue disorganization, and increased extracellular space tortuosity, all contributing to reduced motion of water (21). Correlations with cellularity have been found for some primary and secondary neoplasms (11,22–26). On the other hand, even when using ADC there is also the difficulty of differential diagnosis of malignant and benign lesion, such as lymph node metastasis and ovarian lesion (27,28). The usefulness of measuring ADC in patients with endometrial carcinoma has been reported in several studies (4,29–31); ADC is useful in differentiating between normal and cancerous tissue of the endometrium and cervix, and in differentiating malignant from benign uterine endometrial cavity lesions. Several authors have also reported correlations between ADC and prognostic parameters of endometrioid carcinoma (3–5,32–35), although the results were controversial.
Purpose: To correlate the apparent diffusion coefficient (ADC) of endometrioid carcinoma with histological tumor grade and degree of myometrial invasion. Materials and Methods: 3T diffusion-weighted (DW) magnetic resonance (MR) images of 63 patients were retrospectively reviewed. Two readers measured tumor ADC according to a freehand region of interest (ROI) and a round ROI. Mean and minimum ADCs were correlated with prognostic parameters. Results: The minimum ADC was 0.64 103mm2/s for grade 1 (G1, n ¼ 42), 0.62 for grade 2 (G2, n ¼ 14), 0.46 for grade 3 (G3, n ¼ 7) on freehand ROI. There were significant differences between G1 and G3 (P ¼ 0.007), and G2 and G3 (P ¼ 0.038). No significant correlation was found between tumor grade and mean ADC (0.85 for G1, 0.82 for G2, and 0.72 for G3, P ¼ 0.166). The minimum ADC was significantly lower for patients with deep (n ¼ 21, 0.54) than for those with superficial (n ¼ 39, 0.65) myometrial invasion. Conversely, mean ADC did not differ significantly (0.84 for superficial and 0.78 for deep myometrial invasion, P ¼ 0.081). The same tendency was shown on round ROI. Conclusion: The minimum ADC correlates with prognostic parameters of endometrial carcinoma more strongly than mean ADC. Lower minimum ADC is associated with higher histological tumor grade and higher degree of myometrial invasion. Key Words: MR; diffusion; apparent diffusion coefficient; ADC; endometrioid carcinoma; prognostic parameters J. Magn. Reson. Imaging 2013;00:000–000. C 2013 Wiley Periodicals, Inc. V
1 Division of Radiology, Department of Pathophysiological and Therapeutic Science, Faculty of Medicine, Tottori University, Yonago, Japan. 2 Division of Reproductive-Perinatal Medicine and Gynecological Oncology, Department of Surgery, Faculty of Medicine, Tottori University, Yonago, Japan. *Address reprint requests to: C.I., Division of Radiology, Department of Pathophysiological and Therapeutic Science, Faculty of Medicine, Tottori University, 36-1, Nishi-cho, Yonago, Tottori 683–8504, Japan. E-mail: [email protected] Received August 26, 2013; Accepted November 14, 2013. DOI 10.1002/jmri.24534 View this article online at wileyonlinelibrary.com. C 2013 Wiley Periodicals, Inc. V
1
2
Inoue et al.
Figure 1. ADC measurements of the tumor were performed on an ADC map according to freehand ROI (a) and round ROI (b). Tumor was defined as a focal mass showing high signal intensity on axial DW image (c) compared with the signal of adjacent tissue. The ROIs were carefully set not to involve necrosis by referring to T2-weighted image (d) and contrast enhanced T1-weighted image (e).
The purpose of this study was to examine the correlations between ADC and prognostic parameters of endometrioid carcinoma such as histological grade, degree of myometrial invasion, and the presence of lymph node metastases. MATERIALS AND METHODS Patient Population An initial total of 73 patients were studied retrospectively. All patients had undergone MR examinations including DWI using a 3.0-T MR system between August 2006 and December 2011 at our institution and had received pathological confirmation of endometrial carcinoma. Ten of these patients were excluded because the tumors were too small to be clearly demonstrated on DWI. As a result, a final total of 63 patients (age range, 32–92 years; mean, 60 years) were enrolled. Of the 63 endometrioid adenocarcinomas, 42 were well-differentiated (grade 1), 14 were moderately differentiated (grade 2), and seven were poorly differentiated (grade 3). Three patients did not receive surgical therapy because of distant metastases and pathological confirmation was obtained by biopsy. Pelvic lymph node resection was not performed on seven patients because of a clinical state such as advanced age or preexisting disease. The Institutional Review Board approved the study protocols and waived the requirement for informed consent.
weighted spoiled gradient recalled (SPGR) images (repetition time [TR], 250 msec; echo time [TE], 2.1 msec; flip angle [FA], 75 ; acquisition time, 1 min 30 sec), axial and sagittal T2-weighted fast spin echo (FSE) images (TR, 6500 msec; TE, 100 msec; section thickness, 5 mm; intersection gap, 1.5 mm; acquisition time, 3–3.5 min), and axial and/or sagittal SPGR images with fat suppression (TR, 320 msec; TE, 2.1 msec; FA, 75 ; acquisition time, 1 min 50 sec). Axial DW images were then obtained (TR, 5500–6000 msec; TE, 60–62 msec; inversion time, 200 msec; b factors, 0 and 1000 s/mm2; 112 128 matrix; field of view, 400 mm; section thickness, 4–6 mm with no gap; SENSE reduction factor, 2; signals acquired, 4; acquisition time, 3.5 min). STIR was used for fat suppression in a single spin echo planar imaging sequence, with free breathing during acquisition. Motion-probing gradient pulses were placed in the three orthogonal planes. Isotropic DWI was generated using three orthogonal-axis images. We then obtained axial FSE images as reference images. Parameters for FSE with parallel imaging were as follows: TR, 4500–5000 msec; TE, 100 msec; FA, 160 ; echo spacing, 7.4 msec; echo train length, 20; 512 256 matrix; section thickness, 4–6 mm with no gap; number of excitations, 0.5; free breathing; acquisition time, 50–60 sec. Before imaging, all patients received intramuscular administration of 20 mg of butyl-scopolamine (Buscopan; Nippon Boehringer Ingelheim, Tokyo, Japan) to prevent peristalsis artifacts, except when contraindicated.
MR Technique MRI was performed using a 3.0-T MR system (Signa EXCITE HD; GE Medical Systems, Milwaukee, WI) with phased array coils. Routine pelvic MR images were acquired as follows: axial and sagittal T1-
Image Analysis MR images were independently analyzed by two radiologists (C.I. and S.K.; 8 and 6 years of experience in
Correlation of ADC With Prognostic Parameters
3
Table 1 Mean ADC of Endometrioid Carcinoma With Various Prognostic Factors n
Freehand ROI
Round ROI
G1 G2 G3
