SuperSonic Imagine unveils results of largest ever breast ultrasound
clinical trial
31 May 2010
French company SuperSonic Imagine, has unveiled some of the
results of the largest clinical breast study ever undertaken in
ultrasound imaging. The worldwide multicenter study is assessing the
clinical benefits of ShearWave Elastography in the ultrasonic evaluation
of breast lesions.
The study has two objectives: The first is to demonstrate that
images obtained using ShearWave Elastography are reproducible. The
second is to compare ultrasound alone versus the combination of
ultrasound and ShearWave Elastography for breast lesion diagnosis.
The goal of the latter is to improve lesion classification in
categories BI-RADS 3 and BI-RADS 4 to better direct patients towards
clinical follow-up or biopsy.
“This clinical investigation is the largest trial ever undertaken
by an ultrasound imaging company as the recruitment will involve a
targeted 2300 breast lesion cases,” explained Claude Cohen-Bacrie,
co-founder and Scientific Director of SuperSonic Imagine.
“Today it is essential to obtain additional information on breast
lesions to improve diagnosis. In an era of healthcare reform, being
able to reduce the number of biopsies by correctly classifying
lesions could save resources and spare women the anxiety and
difficulty that surrounds invasive procedures. Better lesion
classification also means improved diagnosis, which can lead to
quicker and better treatment.”
ShearWave Elastography: the technology
Ultrasound imaging plays an important role in breast diagnosis.
It is used on palpable masses, as a second intention exam after
X-ray and MRI and as a modality of choice to guide biopsy. ShearWave
Elastography is a breakthrough technology that gives additional,
important quantitative information about tissue elasticity. Unlike
conventional elastography methods, which rely on manual compression
and measure tissue displacement, it requires no manual compression
and computes true tissue elasticity by measuring the velocity of
shear waves as they propagate in tissue.
Shear wave propagation speed in tissue is directly related to
tissue stiffness. This technology relies upon the generation of a
shear wave and its subsequent capture. Shear wave propagation speed
is then calculated and a colour-coded real time ShearWave
Elastography map is produced showing quantitative, local tissue
stiffness. Results are real-time, user-skill independent,
reproducible and quantifiable in kilopascals.
The multicenter clinical trial
A worldwide multicenter breast clinical study was launched in
April 2008 with 17 American and European sites including: the
Hammersmith Hospital Imperial College (United Kingdom), the Curie
Institute of Paris (France), the DKD Wiesbaden and the academic
hospitals Schleswig-Holstein and Greifswald (Germany), Yale Medical
Center and the Northwestern Memorial Hospital (USA). The study was
conducted under the leadership of Professor David Cosgrove (Imperial
College, London).
The first phase of the study was to define a scientific model on
1000 cases, to determine if ShearWave Elastography information can
complement ultrasound information in order to improve a diagnosis.
To undertake this, it was necessary to identify the criteria of an
elastography image that would, when added to ultrasound criteria,
improve lesion characterization (in sensitivity and specificity when
compared to ultrasound alone) and eventually improve a BI-RADS
score.
The criteria or features studied for each lesion were: size,
shape, average value of elasticity, homogeneity, orientation and
contrast of elasticity between lesion and fatty tissue. The second
phase of the study will consist of testing this scientific model on
an independent set of lesions. This phase is currently ongoing.
Clinical Result 1
ShearWave Elastography features are reproducible
To
determine reproducibility, each clinical investigator was asked to
perform and compare features on 3 separate image acquisitions of the
same lesion. The clinical results clearly showed that it is
reproducible both qualitatively and quantitatively:
- Qualitative: 87% of consecutive repeated exams were
‘similar’, ‘reasonably similar’ or ‘very similar’ in appearance.
- Quantitative: Intra-Observer Reproducibility (IOR)
rates for ShearWave Elastography measurements are close to
perfect at 0.91.
Reproducibility assures the physician of a reliable and precise
evaluation of a lesion, both during an elastography examination and
over time, which is key for follow up.
Clinical Result 2
ShearWave Elastography feature(s) increase diagnostic accuracy
and improve lesion classification
Regression models based on 1000 cases were statistically
evaluated by Caroline Doré, an independent biostatistician from the
Hammersmith Hospital and show that each individual ShearWave
elastographic feature, when added to the ultrasound evaluation,
improved the classification of the BI-RADS score of a lesion. The
global evaluation is calculated on an analysis of the area under the
ROC curve (Receiver Operating Characteristic). The larger the area
under the curve, the better the BI-RADS score classification.
The clinical results demonstrated that when two ShearWave
Elastography features are added to the ultrasound evaluation,
correct breast lesion classification rates soar to 87%, thus leading
to more accurate results.
Scientifically evaluated, the results of this clinical study
therefore demonstrate that ShearWave Elastography features, when
added to the BI-RADS score significantly improve the specificity and
sensitivity of the diagnosis of the lesion. Associated with the
BI-RADS score, these features increase the percentage of correctly
classified lesions and improve lesion diagnosis.
“Historically, ultrasound imaging was considered as an efficient
method to differentiate solid lesions from liquid lesions. As a
result of research over the last 15 years, ultrasound has become an
important technique, with a very high negative predictive value, in
the classification of lesions on the BI-RADS scale: 2( benign) to 5
(highly suggestive of malignancy).
"Today, the results of this multicenter study show that ShearWave
Elastography combined with ultrasound, further improves lesion
classification by significantly raising the percentage of lesions
that are correctly classified and increases the specificity in the
diagnosis while keeping a high negative predictive value and
sensitivity.”
“Our clinical objective is to confirm if ShearWave Elastography,
combined with ultrasound, leads to a more refined lesion
classification of BI-RADS 3 and 4 and in turn leads to a better
direct patients towards follow-up or biopsy,” concludes Claude
Cohen-Bacrie.
BI-RADS is the acronym for Breast Imaging-Reporting and Data
System, a quality control system developed by the ACR (American
College of Radiology) to assess breast lesions according to their
degree of malignancy.