Reduced field-of-view diffusion-weighted MR imaging for assessing the local extent of uterine cervical cancer

Abstract

Background

Recently, the evaluation of the tumor size and local extension of early-stage uterine cervical cancer on magnetic resonance imaging is important for the accurate clinical staging and to determine the indication of less extensive surgery such as fertility sparing radical trachelectomy.

Purpose

To compare the diagnostic ability of reduced field-of-view diffusion-weighted imaging with those of three-dimensional (3D) contrast-enhanced T1-weighted imaging and T2-weighted imaging for assessing the tumor margin delineation and local extent of uterine cervical cancer.

Material and Methods

3T magnetic resonance images, including T2-weighted imaging, reduced field-of-view diffusion-weighted imaging, and 3D contrast-enhanced T1-weighted imaging, in 27 women with surgically proven cervical cancer (19 FIGO stage IB1, 3 IB2, and 5 IIA1) were retrospectively evaluated. Tumor margins and local tumor extent, including the presence of invasion to parametrium and vagina were evaluated on both sagittal and oblique axial (short axis) images; the results were compared with histologically confirmed tumor extension.

Results

Reduced field-of-view diffusion-weighted imaging diagnosed the tumor margins, which was more accurate than T2-weighted imaging (P<0.001) and slightly better than 3D contrast-enhanced T1-weighted imaging. Reduced field-of-view diffusion-weighted imaging could define the tumor margins well even in small lesions (≤ 20 mm). Histological examination revealed parametrial invasion in two cases (clinically under-staged) and vaginal invasion in four cases. Reduced field-of-view diffusion-weighted imaging could demonstrate local extension of all lesions, which was more accurate than clinical examination and T2-weighted imaging.

Conclusion

Addition of reduced field-of-view diffusion-weighted imaging may improve the staging accuracy of magnetic resonance imaging for cervical cancer in assessing the local tumor extent.

Keywords

Genital magnetic resonance diffusion/perfusion magnetic resonance imaging uterus adults primary neoplasms

Introduction

Uterine cervical cancer is the fourth most common cancer in women worldwide and is caused by persistent infection with human papillomavirus (1). Treatment of cervical cancer is stratified by clinical stage according to the International Federation of Gynecology and Obstetrics (FIGO) staging system as delineated in the guidelines (2,3). Magnetic resonance imaging (MRI) is a powerful tool for assessing the tumor size and extension into parametrium, vagina, uterine corpus, or adjacent organs (4 –7). FIGO staging has been based mainly on clinical examination; however, the revised FIGO staging in 2018 allows imaging findings to assign the clinical stage (2,3). Because FIGO 2018 stage IB is divided into stages IB1, IB2, and IB3 based on the size of the tumor—stage IB1< 2 cm, 2 cm ≤ IB2 < 4 cm, 4 cm ≤ IB3—diagnosing tumors < 2 cm is important for the correct staging. In addition, a tumor size ≤ 2 cm may be the indication for less extensive surgery such as fertility-sparing radical trachelectomy (3). Diffusion-weighted imaging (DWI) may demonstrate cervical cancer as a high signal intensity mass with low apparent diffusion coefficient (ADC) value (7 –9); however, detailed evaluation of the local tumor extension such as parametrial or vaginal invasion on conventional DWI is occasionally difficult due to low spatial resolution and distortion caused by susceptibility artifacts. Reduced phase direction field-of-view (FOV) technique by using spatially selective phase encoding gradient can offer high-quality DWI with improved spatial resolution, without phase wrap artifact, and with reduced artifacts related to susceptibility which may be commonly observed in conventional DWI with a larger FOV (10 –15). Hwang et al. evaluated the qualitative image quality of reduced FOV DWI in patients with cervical cancer; reduced FOV DWI achieved better anatomic detail, lesion conspicuity with fewer artifacts, and overall image quality compared to conventional DWI (12).

The purpose of this study is to compare the diagnostic ability of reduced FOV DWI with those of T2-weighted (T2W) imaging and three-dimensional contrast-enhanced T1-weighted (3D CE T1W) imaging including 3D dynamic contrast-enhanced (DCE) MRI in evaluating the tumor margin delineation and local extension of cervical cancer.

Material and Methods

Study population

The institutional review board in our hospital approved this retrospective study and waived the requirement for written informed consent from patients. We cross-referenced the database of the Department of Obstetrics and Gynecology to identify all patients with surgically proven cervical cancer who had undergone MR examinations including reduced FOV DWI between March 2015 and September 2018. A total of 27 women with a mean age of 46 years (age range = 26–67 years) were included in the current study. The histopathological diagnoses were squamous cell carcinomas (n = 19), adenocarcinomas (n = 7), and adenosquamous carcinoma (n = 1). The clinical stages of the patients according to the FIGO 2009 were as follows: stage IB1 = 19 patients; stage IB2 = three patients; stage IIA1 = five patients. The median duration between MRI and surgery was 36 days (range = 22–69 days).

Imaging protocol

MR images of the patients were obtained by using 3T superconducting units (Discovery MR750, GE Healthcare, Waukesha, WI, USA) with 32-channel body-array torso coils. In each examination, T2W imaging and reduced FOV DWI (in all 27 patients), and 3D CE T1W imaging including 3D DCE MRI (in 23/27 patients) were acquired in the sagittal and oblique axial (short axis of the cervix) planes. Fast spin-echo T2W imaging was obtained in the sagittal plane with the following parameters: TR/TE = 5000/97.2–104.3 ms; matrix = 352 × 256; number of acquired signals = 2; FOV = 28 cm; section thickness/section gap = 3/0.5 mm; and in the oblique axial plane which was perpendicular to the cervix as short axis plane with the following parameters: TR/TE = 5000/97–103.6 ms; matrix = 352 × 256; number of acquired signals = 2; FOV = 28–32 cm; section thickness/section gap = 2/0.5–1 mm. Reduced FOV DWI (FOV optimized and constrained undistorted single shot: FOCUS) of the uterus were acquired in the sagittal and oblique axial planes with the following parameters: TR/TE = 4000/55.1–56.3 ms; matrix = 160 × 80; number of signals acquired = 10; FOV = 20–24 × 10–12 cm; section thickness/section gap = 4–5/0 mm; b = 50 and 800 s/mm ². A total of 16 slices of reduced FOV DWI were acquired in 2 min 52 s. 3D DCE MRI was performed after administration of a gadolinium chelate (0.1 mmol/kg of body weight) by using a 3D gradient recalled echo T1W liver acquisition volume acceleration sequence (LAVA) with the following parameters: TR/TE = 4.6/2.1 ms; number of acquired signals = 0.7; matrix = 320 × 192; FOV = 30 cm; section thickness = 3 mm; interpolated to 1.5 mm. Images were acquired at multiple phases of contrast medium enhancement in sagittal plane (pre-contrast and post-contrast at 30, 50, 70, 90, and 110 s). After DCE MRI, 3D CE T1W imaging (LAVA) was obtained in the oblique axial plane with the following parameters: TR/TE = 4–4.1/1.7 ms; number of acquired signals = 0.7; matrix = 288 × 224; FOV = 34–36 cm; section thickness = 3 mm, interpolated to 1.5 mm.

Image analysis

Tumor margins and local tumor extent were visually evaluated on both sagittal and oblique axial (short axis) images with the consensus of two readers (radiologists with 26 and 17 years of experience in body MRI) after careful individual evaluation. The two readers were blinded to the clinical stages and histopathological results but were aware that each patient had cervical cancer. The concordance rate of the two readers was examined by kappa statistic. Diagnostic criteria for cervical cancer included high signal intensity lesion on T2W imaging, which revealed high signal intensity on reduced FOV DWI with low signal intensity on the corresponding ADC map, and low signal intensity on 3D CE T1W imaging. Tumor margins were rated on a three-point scale as: 1 = obscure; 2 = indistinct; and 3 = distinct. Mann–Whitney U test was used to compare the scores between T2W imaging, 3D CE T1W imaging, and reduced FOV DWI. Evaluation of tumor extension included the presence of invasion to parametrium and vagina, and the results were compared with the histologically confirmed tumor extent to determine the diagnostic accuracy of T2W imaging, reduced FOV DWI (together with the corresponding ADC map), and CE 3D T1W imaging including DCE MRI, respectively. The parametrial invasion was regarded as present if nodular or irregular protrusion of the tumor into the parametrium was observed on each sequence. If the low intensity of the vaginal wall was interrupted by the tumor on T2W imaging, or abnormal contrast-enhancement on 3D CE T1W imaging or high signal intensity on reduced FOV DWI exhibiting low signal intensity on the corresponding ADC map continued from the tumor to the vaginal wall, the tumor was considered to invade the vagina.

Results

Mean maximum tumor diameter measured on pathological specimen was 28 mm (range = 5–47 mm) and 10 lesions were relatively small (≤ 20 mm). Tumor margin scores by MRI type are shown in Table 1. Reduced FOV DWI diagnosed the tumor margins, which was more accurate than T2W imaging and CE 3D T1W imaging. Concordance rates between the two readers were 0.78 for T2W imaging, 0.87 for CE 3D T1W imaging, and 1.0 for reduced FOV DWI. The score of reduced FOV DWI was significantly better than that of T2W imaging (P < 0.001) and better than, but not significantly different from, that of 3D CE T1W imaging (P = 0.18) (Fig. 1). The score of 3D CE T1W imaging was also significantly better than that of T2W imaging (P < 0.005). Reduced FOV DWI could demonstrate even small lesions (≤ 20 mm, n = 10) as high intensity areas with distinct tumor margin (Fig. 2). Reduced FOV DWI also diagnosed the tumor margins; this was more accurate than T2W imaging and 3D CE T1W imaging in small lesions (≤ 20 mm) (Table 2).

Fig. 1.

A 51-year-old woman with cervical cancer (stage IB2, squamous cell carcinoma). (a) Sagittal fast spin-echo T2W imaging shows an irregular cervical mass with obscure margin (arrow). (b) Sagittal 3D CE T1W imaging with fat saturation (LAVA) shows the tumor (arrow) as a low intensity lesion with distinct margin. (c) Sagittal reduced FOV DWI shows the tumor (arrow) as a high intensity lesion with distinct margin. (d) A photomicrograph at low magnification shows tumor cells infiltrating myometrium.T2W: T2-weighted; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted; LAVA: liver acquisition volume acceleration sequence; FOV DWI: field-of-view diffusion-weighted imaging.

Fig. 2.

A 44-year-old woman with small cervical cancer (stage IB1, squamous cell carcinoma, pathologically confirmed maximum diameter is 20 mm). (a) Sagittal fast spin-echo T2W imaging shows a cervical mass with obscure margin (arrow). (b) Sagittal 3D CE T1W imaging with fat saturation (LAVA) shows the tumor (arrow) as a low intensity lesion with distinct margin. (c) Sagittal reduced FOV DWI shows the tumor (arrow) as a high intensity lesion with distinct margin. (d) Sagittal ADC map corresponding to the reduced FOV DWI shows the tumor (arrow) as a low ADC lesion. (e) Photograph of resected uterus reveals a small cervical mass (arrow).T2W: T2-weighted; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted; LAVA: liver acquisition volume acceleration sequence; FOV DWI: field-of-view diffusion-weighted imaging; ADC: apparent diffusion coefficient.

Table 1.

Tumor margin scores by MRI.

	Score 1 (obscure)	Score 2 (indistinct)	Score 3 (distinct)	Mean score
T2W imaging (n = 27)	7	16	4	1.9
Reduced FOV DWI (n = 27)	2^*	0	25	2.9
3D CE T1W imaging (n = 23)	4	3	16	2.5

*Due to severe susceptibility artifact. T2W: T2-weighted; FOV DWI: field-of-view diffusion-weighted imaging; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted.

Table 2.

Tumor margin scores by MRI (≤ 20 mm).

	Score 1 (obscure)	Score 2 (indistinct)	Score 3 (distinct)	Mean score
T2W imaging (n = 10)	5	3	2	1.7
Reduced FOV DWI (n = 10)	2^*	0	8	2.6
3D CE T1W imaging (n = 10)	4	2	4	2.0

*Due to severe susceptibility artifact.T2W: T2-weighted; FOV DWI: field-of-view diffusion-weighted imaging; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted.

Histological examination revealed parametrial invasion in 2/27 cases (clinically under-staged); each of the three MR sequences could demonstrate parametrial invasion correctly in these two cases (Table 3) (Fig. 3). The concordance rate between the two readers in evaluating parametrial invasion was 1.0 for all three sequences. On the other hand, vaginal invasion was revealed histologically in 4/27 cases (clinically, one of four cases with vaginal invasion was under-staged and two of the other 23 cases without vaginal invasion were over-staged); reduced FOV DWI could demonstrate vaginal invasion correctly in all four cases. 3D CE T1W imaging was performed in 3/4 cases with vaginal invasion and could demonstrate vaginal invasion correctly in all three cases, whereas T2W imaging underestimated one of the four cases with vaginal invasion (Fig. 4). The concordance rate between the two readers in evaluating vaginal invasion was 1.0 for all three sequences. The presence or absence of parametrial invasion as documented by reduced FOV DWI (n = 25, two patients with suboptimal images due to severe susceptibility artifacts were excluded), T2W imaging (n = 27), and 3D CE T1W imaging (n = 23) agreed with the histopathological findings in all 25 patients (100%) on reduced FOV DWI, all 27 patients (100%) on T2W imaging, and all 23 patients (100%) on 3D CE T1W imaging. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of reduced FOV DWI, T2W imaging, and 3D CE T1W imaging for assessing parametrial invasion are all 100%. The presence or absence of vaginal invasion as documented by reduced FOV DWI, T2W imaging, and 3D CE T1W imaging agreed with the histopathological findings in all 25 patients (100%) on reduced FOV DWI, 26/27 patients (96%) on T2W imaging, and all 23 patients (100%) on 3D CE T1W imaging. The sensitivity, specificity, PPV, and NPV of reduced FOV DWI and CE 3D T1W imaging for assessing vaginal invasion are all 100%, whereas those of T2W imaging are 75%, 100%, 100%, and 96%, respectively.

Fig. 3.

A 42-year-old woman with cervical cancer (stage IB1, squamous cell carcinoma). Pathological examination after surgery revealed the right parametrial invasion. (a) Oblique axial fast spin-echo T2W imaging shows a cervical mass with indistinct margin (arrow). Low intensity stromal ring is interrupted at the right side with right parametrial tumor invasion. (b) Oblique axial 3D CE T1W imaging with fat saturation (LAVA) shows the tumor with right parametrial invasion (arrow) as a low intensity lesion with distinct margin. (c) Oblique axial reduced FOV DWI shows the tumor with right parametrial invasion (arrow) as a high intensity lesion with distinct margin. (d) Photograph of resected uterus reveals a cervical mass with parametrial invasion (arrow).T2W: T2-weighted; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted; LAVA: liver acquisition volume acceleration sequence; FOV DWI: field-of-view diffusion-weighted imaging.

Fig. 4.

A 51-year-old woman with cervical cancer (stage IB1, squamous cell carcinoma). Pathological examination after surgery revealed vaginal invasion. (a) Sagittal fast spin-echo T2W imaging shows a cervical mass with indistinct margin (arrow). (b) Sagittal 3D CE T1W imaging with fat saturation (LAVA) shows the tumor (arrow) with vaginal invasion (arrowheads) as a low intensity lesion with distinct margin. (c) Sagittal reduced FOV DWI shows the tumor (arrow) with vaginal invasion (arrowheads) as a high intensity lesion with distinct margin. (d) Sagittal ADC map corresponding to the reduced FOV DWI shows the tumor (arrow) with vaginal invasion (arrowheads) as a low ADC lesion. (e) Photograph of resected uterus reveals a cervical mass with vaginal invasion (arrow).T2W: T2-weighted; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted; LAVA: liver acquisition volume acceleration sequence; FOV DWI: field-of-view diffusion-weighted imaging; ADC: apparent diffusion coefficient.

Table 3.

Local tumor extension evaluated by MRI.

		T2W imaging (n = 27)		Reduced FOV DWI (n = 25)*		3D CE T1W imaging (n = 23)^†
	Pathology (n = 27)	–	+	–	+	–	+
Parametrium –	25	25	0	23	0	21	0
Parametrium +	2	0	2	0	2	0	2
Vagina –	23	23	0	21	0	20	0
Vagina +	4	1	3	0	4	0	3

*Two suboptimal cases were excluded due to severe susceptibility artifact.

^†Four cases without contrast-enhanced study were excluded.T2W: T2-weighted; FOV DWI: field-of-view diffusion-weighted imaging; 3D CE T1W: three-dimensional contrast-enhanced T1-weighted.

Discussion

The FIGO cancer staging system was created in 1958 and revised in 1988, 2009, and 2018. Staging of cervical cancer was initially based upon clinical examination by experienced gynecologist till FIGO 2009; however, optional use of cross-sectional imaging such as computed tomography (CT), MRI and positron emission tomography (PET)/CT was recommended at FIGO 2009 (2,3). In FIGO 2018, imaging findings are allowed to assign the stage where resources are available (3). Surgical treatment is indicated for early stage invasive cervical cancer. Because radical trachelectomy may be indicated for tumors measuring ≤ 2 cm in the largest diameter in young women desiring fertility sparing, the importance of small cancer delineation on MRI is increased (3). T2W imaging is essential for the staging of cervical cancer with MRI according to the guidelines; however, tumors may not always be well demonstrated on T2W imaging and the use of intravenous contrast medium or DWI may be recommended (4 –9). Akita et al. have reported higher detectability of cervical cancer on CE T1W imaging (95%) compared with that on T2W imaging (76%) (16). Park et al. have reported high specificity (96–97%) and accuracy (90%) in assessing parametrial invasion of cervical cancer on fused T2W imaging and DWI (17). Cervical cancer may show high signal intensity on DWI with significantly lower ADC values than normal cervical tissue, facilitating definition of local tumor extent (4,7–9). However, relatively low spatial resolution and susceptibility-induced signal distortion may cause misinterpretation of local extent of cervical cancer on DWI (4,7–9).

The recently introduced reduced FOV DWI technique with high spatial resolution and reduced off-resonance-induced artifacts could enhance the diagnostic ability in the lesion detection and for the evaluation of tumor extent of gynecologic tumors such as endometrial cancer (12 –15). Reduced FOV DWI uses two-dimensional spatially selective echo planar radiofrequency (RF) excitation pulse followed by a 180° refocusing RF pulse to select a rectangular FOV, decreasing the read-out duration by reducing the phase-encode direction within the excited FOV (10,11). In the current study, most lesions showed high signal intensity with distinct margins on reduced FOV DWI, except for two lesions with severe distortion caused by susceptibility artifact from adjacent rectal gas. In particular, even small lesions (≤20 mm) could be demonstrated with distinct tumor margins on reduced FOV DWI compared with T2W imaging.

In assessing the local extent of cervical cancer, reduced FOV DWI could reveal parametrial and vaginal invasion more correctly than clinical examination and T2W imaging. There was no significant difference between reduced FOV DWI and 3D CE T1W imaging in the diagnostic ability for assessing local extension of cervical cancer; however, CE T1W imaging may have contraindications such as severe renal dysfunction at risk for nephrogenic systemic fibrosis, allergy to gadolinium-containing agents, or during pregnancy (14). Since the retrospective nature and relatively small population with heterogeneity in tumor histopathology are limitations to the current study, further prospective studies in larger populations to support the results are needed.

In conclusion, the addition of reduced FOV DWI may improve the staging accuracy of cervical cancer in assessing local tumor extension such as parametrial and vaginal invasion. Reduced FOV DWI could delineate the tumor margins clearly in even small lesions; however, susceptibility artifacts from adjacent intestinal gas may cause suboptimal image quality in seldom cases.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ferlay

Colombet

Soerjomataram

, et al. Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer 2019; 144:1941–1953.

Odicino

Pecorelli

Zigliani

, et al. History of the FIGO cancer staging system. Int J Gynaecol Obstet 2008; 101:205–210.

Bhatla

Aoki

Sharma

, et al. Cancer of the cervix uteri. Int J Gynaecol Obstet 2018; 143 (Suppl. 2):22–36.

Bourgioti

Chatoupis

Moulopoulos

LA.

Current imaging strategies for the evaluation of uterine cervical cancer.

World J Radiol 2016; 8:342–354.

Freeman

Aly

Kataoka

, et al. The revised FIGO staging system for uterine malignancies: implications for MR imaging. Radiographics 2012; 32:1805–1827.

Balleyguier

Sala

Da Cunha

, et al. Staging of uterine cervical cancer with MRI: guidelines of the European Society of Urogenital Radiology. Eur Radiol 2011; 21:1102–1110.

Sala

Rockall

Freeman

, et al. The added role of MR imaging in treatment stratification of patients with gynecologic malignancies: what the radiologist needs to know. Radiology 2013; 266:717–740.

Dhanda

Thakur

Kerkar

, et al. Diffusion-weighted imaging of gynecologic tumors: diagnostic pearls and potential pitfalls. Radiographics 2014; 34:1393–1416.

Dappa

Elger

Hasenburg

, et al. The value of advanced MRI techniques in the assessment of cervical cancer: a review. Insights Imaging 2017; 8:471–481.

10.

Saritas

Cunningham

Lee

, et al. DWI of the spinal cord with reduced FOV single-shot EPI. Magn Reson Med 2008; 60:468–473.

11.

Andre

Bammer

Advanced diffusion-weighted magnetic resonance imaging techniques of the human spinal cord.

Top Magn Reson Imaging 2010; 21:367–378.

12.

Hwang

Hong

Kim

, et al. Reduced field-of-view diffusion-weighted MRI in patients with cervical cancer. Br J Radiol 2018; 91:20170864.

13.

Bhosale

Iyer

, et al. Feasibility of a reduced field-of-view diffusion-weighted (rFOV) sequence in assessment of myometrial invasion in patients with clinical FIGO stage I endometrial cancer. J Magn Reson Imaging 2016; 43:316–324.

14.

Takeuchi M, Matsuzaki K, Harada M. Evaluating myometrial invasion in endometrial cancer: comparison of reduced field-of-view diffusion-weighted imaging and dynamic contrast-enhanced MR imaging. Magn Reson Med Sci 2018;17:28–34.

15.

Ota

Hori

Onishi

, et al. Preoperative staging of endometrial cancer using reduced field-of-view diffusion-weighted imaging: a preliminary study. Eur Radiol 2017; 27:5225–5235.

16.

Akita

Shinmoto

Hayashi

, et al. Comparison of T2-weighted and contrast-enhanced T1-weighted MR imaging at 1.5 T for assessing the local extent of cervical carcinoma. Eur Radiol 2011; 21:1850–1857.

17.

Park

Kim

Park

, et al. Parametrial invasion in cervical cancer: fused T2-weighted imaging and high-b-value diffusion-weighted imaging with background body signal suppression at 3 T. Radiology 2015; 274:734–741.