Quantitative serial MRI of the treated fibroid uterus

Abstract

Objective: There are no long-term medical treatments for uterine fibroids, and non-invasive biomarkers are needed to evaluate novel therapeutic interventions. The aim of this study was to determine whether serial dynamic contrast-enhanced MRI (DCE-MRI) and magnetization transfer MRI (MT-MRI) are able to detect changes that accompany volume reduction in patients administered GnRH analogue drugs, a treatment which is known to reduce fibroid volume and perfusion. Our secondary aim was to determine whether rapid suppression of ovarian activity by combining GnRH agonist and antagonist therapies results in faster volume reduction.

Methods: Forty women were assessed for eligibility at gynaecology clinics in the region, of whom thirty premenopausal women scheduled for hysterectomy due to symptomatic fibroids were randomized to three groups, receiving (1) GnRH agonist (Goserelin), (2) GnRH agonist+GnRH antagonist (Goserelin and Cetrorelix) or (3) no treatment. Patients were monitored by serial structural, DCE-MRI and MT-MRI, as well as by ultrasound and serum oestradiol concentration measurements from enrolment to hysterectomy (approximately 3 months).

Results: A volumetric treatment effect assessed by structural MRI occurred by day 14 of treatment (9% median reduction versus 9% increase in untreated women; P = 0.022) and persisted throughout. Reduced fibroid perfusion and permeability assessed by DCE-MRI occurred later and was demonstrable by 2-3 months (43% median reduction versus 20% increase respectively; P = 0.0093). There was no apparent treatment effect by MT-MRI. Effective suppression of oestradiol was associated with early volume reduction at days 14 (P = 0.041) and 28 (P = 0.0061).

Conclusion: DCE-MRI is sensitive to the vascular changes thought to accompany successful GnRH analogue treatment of uterine fibroids and should be considered for use in future mechanism/efficacy studies of proposed fibroid drug therapies. GnRH antagonist administration does not appear to accelerate volume reduction, though our data do support the role of oestradiol suppression in GnRH analogue treatment of fibroids.

Trial registration: ClinicalTrials.gov NCT00746031.

Conflict of interest statement

Competing Interests: HODC is a PLOS ONE Editorial Board member (since 2012 an Academic Editor). She confirms that this does not alter her adherence to all the PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Sagittal (a) and axial-oblique (b) T2-weighted images showing dimensions of a large fibroid.

(c) Magnetization transfer-weighted and (d) dynamic contrast-enhanced MRI (DCE-MRI) image (acquired one minute after injection of contrast) of the same patient.

Figure 2. CONSORT flow diagram showing enrolment and progress of study participants.

Figure 3. Box and whisker plot showing blood oestradiol concentration, with groups 1, 2 and 3 displayed in black, red and blue respectively. *

indicates samples taken on MRI scan days.

Figure 4. Median % volume change of (a) uterus and (b) largest fibroid from baseline at day 14, day 28 and 2 to 3 months (i.e. within 10 days of hysterectomy), measured by T2-weighted MRI; error bars show the interquartile range.

Figure shows data for treated (groups 1 and 2) and untreated participants.

Figure 5. Changes in uterine vascular properties at day 14, day 28 and 2 to 3 months (i.e. within 10 days of hysterectomy).

(a) Shows median % change in the fibroid dynamic contrast-enhanced MRI (DCE-MRI) parameter kep from baseline, while (b) shows corresponding changes in pulsatility index (PI) as measured by Doppler ultrasound of the uterine arteries. Error bars show the interquartile range. Figure shows data for treated (groups 1 and 2) and untreated participants.

Figure 6. Relationship between oestradiol concentration, and volumetric and vascular changes.

(a) shows median uterine volume change, assessed by T2-weighted MRI; error bars show the interquartile range. Data are shown for treated participants (groups 1 and 2), classified as oestradiol-suppressed and -unsuppressed as described in the text. (b) shows changes in uterine volume and fibroid kep (DCE-MRI) for all patients (group 1: black, group 2: red, group 3: blue) at the final MRI scan. The areas of the data points are proportional to blood oestradiol concentration at the same time point (the largest data point corresponds to a value of 1153 pg/mL).