Sonata system for diagnostic imaging and treatment of symptomatic uterine fibroids

Study size, design and location

Subgroup analysis of 197 people (534 treated fibroids) from 2 prospective clinical trials (FAST‐EU and SONATA) who had submucous or large fibroids treated with the Sonata (or VizAblate) system.

Intervention and comparator

Sonata (or VizAblate) system, no comparator.

Key outcomes

In the study, 86% of people with submucous fibroids only and 81% of people with large fibroids (over 5 cm in diameter) experienced bleeding reduction within 3 months after treatment with the Sonata system. During the 12 months after the procedure, overall symptom severity and health‐related quality of life showed sustained, significant improvements. MRI imaging of fibroids in the FAST-EU trial showed an average volume reduction of 68%. Among people with submucous fibroids only, the rate of surgical reintervention during 12 months of follow‐up was 3.7% in the FAST‐EU trial and 0% in the SONATA trial.

Strengths and limitations

The FAST-EU and SONATA studies are multicentre studies with prospective design. The FAST-EU trial included 2 UK centres, increasing generalisability to the NHS (although the UK cohort had a small sample size).

Both studies were single-arm, so direct comparisons with alternative treatment options cannot be made. Both studies were funded by the company.

Study size, design and location

Prospective, multicentre, single-arm cohort study (the SONATA trial) of 147 women before the menopause (aged between 25 and 50) with symptomatic uterine fibroids who elected for transcervical fibroid ablation (TFA) treatment. Lukes et al. (2020) reported on the 3-year outcomes, Miller et al. (2019) on the 2‑year outcomes and Chudnoff et al. (2019) on the 12-month outcomes.

The study included 22 investigational centres in the US (n=21) and Mexico (n=1). It was done under an investigational device exemption from the US Food and Drug Administration.

Intervention and comparator

Sonata system, no comparator.

Key outcomes

Four women were excluded from the full efficacy analysis set at 12 months because they reached menopause and were unable to provide a pictorial blood loss assessment chart diary at their 12-month visits. The study met its co-primary endpoints at 12 months (n=143, full analysis set), with 64.8% of women experiencing 50% or greater reduction in menstrual bleeding and 99.3% of women who did not have surgical intervention for heavy menstrual bleeding. The mean pictorial blood loss assessment chart score decreased by 38.9%, 48.4%, and 51.1% at 3, 6, and 12 months after the procedure, respectively (p<0.001). At 12 months, 95.1% of women had experienced a reduction in menstrual bleeding. The mean maximal reduction in fibroid volume per patient was 62.4% (p<0.001). Symptom severity and health-related quality of life scores significantly improved by 32.1 points and 43.7 points, respectively, at 12 months (p<0.001). On average, women reported returning to normal activity in 2.2 days, with more than half returning to normal activity within 1 day. There were 97% of women who were satisfied with treatment and 96.3% noted improvements in their fibroid symptoms at 12 months. No device-related adverse events (AEs) were reported. Two serious procedural-related AEs were reported in 2 women (1.4%) and nonserious procedure-related AEs were reported in 74 women (50.3%).

There were 85% of women enrolled who returned for follow up at 2 years. Compared with baseline, symptom severity statistically decreased from 55 (plus or minus 19) to 24 (plus or minus 18; p<0.001), health-related quality of life statistically increased from 40 (plus or minus 21) to 83 (plus or minus 19; p<0.001). EQ-5D scores statistically increased from 0.72 (plus or minus 0.21) to 0.89 (plus or minus 0.14; p<0.001). Over 2 years, surgical reintervention for heavy menstrual bleeding was done in 5.5% of women. There was 1 pregnancy with a normal peripartum outcome. Indicators of work impairment because of fibroid symptoms significantly improved from baseline to 2 years (including missed work time, overall work impairment, and activity impairment). There were 94% of women who were satisfied with treatment at 2 years (75% very satisfied, 13% moderately satisfied and 6% somewhat satisfied).

There were 15 women lost to follow up during the trial, with 90% of women accounted for at 3 years (132 people out of 147). Compared with baseline, mean symptom severity scores decreased from 55 (plus or minus 19) to 22 (plus or minus 21), health-related quality of life increased from 40 (plus or minus 21) to 83 (plus or minus 23). EQ-5D increased from 0.72 (plus or minus 0.21) to 0.88 (plus or minus 0.16; all statistically significant with p<0.001). The 3-year rates of surgical reintervention for heavy menstrual bleeding calculated by the binomial and Kaplan–Meier methods were 9.2% and 8.2%, respectively. Surgical reinterventions included 10 hysterectomies and 1 endometrial ablation. All work productivity and activity level parameters improved significantly during the 3-year follow up. At 3 years, 94% of women were satisfied with treatment (71% very satisfied, 14% moderately satisfied, 9% somewhat satisfied).

Strengths and limitations

This was a multicentre study with prospective design and endpoints defined using set objective performance criteria. A relatively robust patient selection process may have helped to minimise confounding factors. The study included patient-reported outcomes.

This was a single-arm study, so direct comparisons with alternative treatment options cannot be made. The study was funded by the company and did not involve any UK centres, limiting its generalisability to the NHS. Treatment was limited to fibroids that were between 1 cm and 5 cm in diameter. Women who wanted to become pregnant in the future were excluded.

Study size, design and location

Prospective, multicentre, single-arm cohort study (the FAST-EU trial) of 50 women (aged 28 or older) with uterine fibroids and heavy menstrual bleeding. The study was done across 7 centres in Mexico (n=1), the Netherlands (n=4) and UK (n=2). Brölmann et al. (2016) reported on 12-month clinical outcomes and Huirne et al. (2018) reported on health utility scores during the 12 months after treatment.

Intervention and comparator

VizAblate system, no comparator.

Key outcomes

There were 50 women (89 fibroids) who had treatment with the VizAblate system. One woman was excluded from primary efficacy analysis because of unusable imaging results. At 3 and 12 months, perfused fibroid volumes were reduced from baseline by an average of 68% (49 women; 89 fibroids) and 67% (28 women; 43 fibroids), respectively. Total fibroid volumes were reduced from baseline by an average of 55% (49 women; 89 fibroids) and 67% (28 women; 43 fibroids), respectively (p<0.001 for all compared with baseline). At 12 months, mean menstrual pictogram scores and symptom severity scores decreased by 54% (n=48; p<0.001) and 55% (n=49; p<0.001), respectively. The mean health-related quality of life (HRQoL) scores increased by 277% (n=48; p<0.001). There were 4 women (8%) who had surgical reintervention within 12 months; all were 6 months after treatment.

There were 49 women who provided responses to the EQ-5D-3L. Data from 1 woman who conceived 3 months after treatment and gave birth at term was excluded from analysis. Overall, treatment with VizAblate was associated with an increase in health utility scores, from a mean of 0.745 at baseline, to means of 0.838, 0.852, and 0.914 at 3 months, 6 months, and 12 months, respectively. The change from baseline at 12 months was statistically significant (p<0.001). When stratified by country, the improvement in health utility at 12 months was statistically significant for both the cohorts in Mexico (p<0.004) and Holland (p value less than 0.003), but not for the UK cohort (p=0.75). After translating health utility values into quality-adjusted life years (QALYs), the overall cohort experienced an average of 0.85 QALYs in the 1‑year period from baseline to 12-month follow up.

The preservation of uterine wall integrity 12 months after treatment with VizAblate was also assessed for the 29 women using baseline and 12-month MRI image data. This secondary analysis (the INTEGRITY study, Bongers et al. 2019) showed that there were no areas on the MRI that indicated loss of myometrial integrity compared with baseline.

Strengths and limitations

This was a multicentre and international study with 2 centres in the UK, increasing generalisability to the NHS. The study had a prospective design and consecutive recruitment of patients, reducing risk of bias. MRI results were reviewed by an independent core laboratory to reduce variability and bias in MRI image quality and interpretation. The study used a validated, and standardised tool (EQ-5D-3L).

This was a single-arm study, so direct comparisons with alternative treatment options cannot be made. Only 58% of eligible patients (28 people out of 48) had MRI at 12 months. Follow-up time was limited to 1 year, so long-term effects are unclear. Women who wanted to have a baby in the future or had fibroids larger than 5 cm were not included. Although the study included 2 UK centres, the UK cohort had a small sample size (n=6) and half of all missing values in the study were from the UK cohort. The UK cohort may have been underpowered to detect statistical differences in health utility from baseline to 12-month follow up. The effect of socioeconomic status on health utility results cannot be excluded.

Study size, design and location

Retrospective, single-arm, data-collection cohort study (the VITALITY study) involving 23 women aged 28 and over with heavy menstrual bleeding secondary to fibroids. The study was done at a FAST-EU study site in Mexico and included women who had previously been enrolled and treated in the 12‑month FAST-EU trial.

Intervention and comparator

Vizablate system, no comparator.

Key outcomes

The study generated long-term follow-up information from 17 women (73.9%), with a mean follow up of 64.4 months (range 57 to 73 months). The mean symptom severity scores significantly decreased from 64.9 at baseline to 27.6 at long-term follow up (p=0.002). Mean HRQoL scores significantly improved from 27.2 at baseline to 76.0 at long-term follow up (p=0.0001). No surgical reinterventions happened through the first 3.5 years after treatment. Two hysterectomies were reported, 1 after 3.5 years after ablation and another about 4 years after ablation. Overall, there was a 12% incidence of surgical reinterventions over an average of 5.4 years follow up. This corresponds to an event rate of 2.2% per year. Freedom from surgical reintervention at 1, 2, and 3 years was 100%, and, at 4 and 5 years, was 88.2%. One pregnancy happened, and this was within the first year of treatment. It led to a normal-term birth by elective repeat caesarean and has been reported in a previous case report (Garza-Leal et al. 2014).

Strengths and limitations

The study had a high (around 75%) patient long-term follow-up rate. Treatment was given by a single operator, which may have allowed consistent perioperative management among patients. HRQoL scores and symptom severity scores were recorded and used to complement data on surgical reintervention.

The study was limited by having a relatively small patient cohort. Because it was a single-centre study done in Mexico the generalisability of results to an NHS setting are limited. It was a single-arm study, so direct comparisons with alternative treatment options cannot be made.

Study size, design and location

Prospective, multicentre, single-arm cohort study (the OPEN trial) involving 37 people having transcervical radiofrequency ablation treatment for symptomatic fibroids. Everyone had at least 1 FIGO type 1, 2 or 2 to 5 (transmural) uterine fibroid. The presence or absence of intrauterine adhesions was assessed at baseline before treatment with diagnostic hysteroscopy and again at 6 weeks after ablation (second-look hysteroscopy). The study was done at 6 academic and community hospitals in the UK (n=1), the Netherlands (n=1), Switzerland (n=1), and Germany (n=3).

Intervention and comparator

Sonata system, no comparator.

Key outcomes

There were 50 fibroids with a mean diameter of 3.4 (plus or minus 1.8 cm; ranging from 1 cm to 8 cm) ablated. Of the 37 people enrolled, 35 completed the study follow up and 2 electively withdrew from the study before the completion of study follow up. There were 34 out of 35 people who had paired baseline and second-look hysteroscopies that could be evaluated by the independent readers. One person was excluded from the analysis because of an unevaluable hysteroscopy video. At 6 weeks, none of these hysterectomy videos revealed any formation of intrauterine adhesions after treatment with Sonata. This included 6 people with apposing fibroids, that usually have a substantial risk of forming adhesions after operative hysteroscopic treatment.

Strengths and limitations

This was a multicentre study with a prospective design. The study used independent reviewers for the assessment of baseline and post-treatment hysteroscopy videos. One of the investigational sites was in the UK, increasing the generalisability to the NHS.

This was a single-arm study, so direct comparisons with alternative treatment options cannot be made. The study was not intended to be statistically powered and it is not clear how many of the people included in the analysis were treated in the UK. The study was funded by the company and 2 of the authors were consultants for the company.

Study size, design and location

Systematic review and meta-analysis of prospective studies for radiofrequency fibroid ablation (RFA) in 1,283 people (median age: 42 years) with symptomatic uterine fibroids (treated between 2003 and 2018). It included a total of 32 articles: 19 on laparoscopic RFA, 8 on transvaginal RFA and 5 on transcervical RFA (TFA; using Sonata or VizAblate system). Included studies were done in Mexico, US, Canada, Germany, Italy, Korea, Latin America, Norway, Denmark, China, Spain.

Intervention and comparator

Laparoscopic, transvaginal or transcervical RFA.

Key outcomes

There were 19 out of 20 prospective primary studies (reported in 32 articles) that had good or fair methodological quality. Across all 3 types of RFA delivery approaches the mean procedure time was 49 minutes. Procedure time was statistically different among RFA delivery approaches (laparoscopic, 73 minutes; TFA, 44 minutes; transvaginal, 24 minutes), when for all pairwise comparisons p≥0.002. The mean time to discharge, time to normal activities, and time to return to work were 8.2 hours, 5.2 days and 5.1 days, respectively. For TFA the time to discharge, time to normal activities, and time to return to work were 2.5 hours, 3.3 days and 3.6 days, respectively. At 12 months follow up, fibroid volume decreased by 66%, HRQoL increased by 39 points, and symptom severity scores decreased by 42 points (for all p<0.001 compared with baseline). The annual cumulative rate of reinterventions because of fibroid-related symptoms was 4.2%, 8.2%, and 11.5% over 3 years. The reintervention rate at 12 months was comparable among TFA (2.7%), laparoscopic RFA (3.8%), and transvaginal RFA (5.3%), when p≥0.52 for all pairwise comparisons. The conclusions of this meta-analyses were unchanged among several sensitivity analyses.

Strengths and limitations

The conduct, analysis and reporting of the review adhered to the preferred reporting items for systematic reviews and meta-analyses (PRISMA). Only prospective studies were included. The review included a large sample size and conclusions were unchanged in various sensitivity analyses. Among the 20 prospective primary studies in this review, which were reported in 32 articles, study quality was rated as good or fair for 19 of 20 studies.

The company provided funding for the study. There were insufficient studies to statistically compare RFA delivery types for several outcomes and it was not possible to determine whether RFA efficacy was influenced by fibroid type or volume because of concerns of aggregation bias. Only a few of the included studies had a follow up of longer than 12 months.