Breast Neoplasms
Conditions
Keywords
breast, lesion, ultrasound, BI-RADS, malignancy, women
Brief summary
Our hypothesis is that the addition of ShearWave Elastography (SWE) to a conventional breast ultrasound examination provides useful information for the radiologist when imaging lesions in the breast, as compared to conventional grayscale ultrasound alone.
Interventions
None listed
Sponsors
SuperSonic Imagine
Study design
Observational model
CASE_ONLY
Time perspective
CROSS_SECTIONAL
Eligibility
Sex/Gender
FEMALE
Age
21 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* patients who have been referred to a breast ultrasound because of a positive physical palpation and/or positive mammograms and/or positive ultrasound and/or positive MRI. * female * age 21 or older * provide informed consent
Exclusion criteria
* Women who are unwilling or unable to provide informed consent * Women with breast implants * Women with superficial lesions or lesions on skin (with the most superficial surface of the lesion within 5 mm of the skin surface) * Women who are pregnant or breastfeeding * Women who are undergoing chemotherapy or radiotherapy for any cancer * Women with previous breast conserving surgery on the breast of interest Note: Previous excision of a benign lesion 4 cm or more away from the suspected lesion does not constitute an exclusion criterion.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | 2 years | Positive reference standard = malignant cytologic or histopathologic result. Negative reference standard = BIRADS 2 lesions, BIRADS 3 lesions with benign histopathology, or a 1 year follow-up ultrasound exam showing resolved or decreased lesion size. Conservative strategy:features of E-homogeneity, E-max and E-color were used to upgrade BIRADS 3 lesions to BIRADS 4a' or downgrade BIRADS 4a lesions to BIRADS 3'. Aggressive strategy used the same features but upgraded and downgraded more lesions. Based on 939 lesions. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | performed on the same day, within 2 years from study start date | 614 benign and 144 malignant breast lesions were scanned in SWE 3 consecutive times, and the similarity of the 3 images was evaluated by the investigator. |
| Intraobserver Reliability of Quantitative SWE Measurements | performed on the same day, within 2 years from study start date | 614 benign and 144 malignant lesions. Each category's measurement (diameter, area, etc..) is performed 3 times. These 3 measurements are then compared with each other in order to calculate the interclass correlation coefficient. |
| Interobserver Agreement of B Mode Ultrasound and SWE Features | performed on the same day, after study completion | 614 benign and 144 malignant breast masses. |
Countries
France, Germany, Italy, United Kingdom, United States
Participant flow
Recruitment details
Recruitment dates: October 23, 2008 - June 15, 2010 Hospitals and private medical clinics in 16 sites worldwide participated.
Participants by arm
| Arm | Count |
|---|---|
| Breast Lesion B-mode ultrasound examination with currently marketed machine followed by ultrasound B-mode examination and ShearWave Elastography examination | 1,681 |
| Total | 1,681 |
Withdrawals & dropouts
| Period | Reason | FG000 |
|---|---|---|
| Overall Study | Inadequate acquisition or analysis | 50 |
| Overall Study | inclusion/exclusion criteria not met | 25 |
| Overall Study | Lost to Follow-up | 6 |
| Overall Study | no pathology obtained | 10 |
| Overall Study | technical/machine error | 28 |
Baseline characteristics
| Characteristic | Breast Lesion |
|---|---|
| Age Continuous | 51.8 years STANDARD_DEVIATION 14.5 |
| Region of Enrollment France | 322 participants |
| Region of Enrollment Germany | 529 participants |
| Region of Enrollment Italy | 89 participants |
| Region of Enrollment United Kingdom | 175 participants |
| Region of Enrollment United States | 566 participants |
| Sex: Female, Male Female | 1681 Participants |
| Sex: Female, Male Male | 0 Participants |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | — / — |
| other Total, other adverse events | 0 / 0 |
| serious Total, serious adverse events | 0 / 0 |
Outcome results
Primary
Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics
Positive reference standard = malignant cytologic or histopathologic result. Negative reference standard = BIRADS 2 lesions, BIRADS 3 lesions with benign histopathology, or a 1 year follow-up ultrasound exam showing resolved or decreased lesion size. Conservative strategy:features of E-homogeneity, E-max and E-color were used to upgrade BIRADS 3 lesions to BIRADS 4a' or downgrade BIRADS 4a lesions to BIRADS 3'. Aggressive strategy used the same features but upgraded and downgraded more lesions. Based on 939 lesions.
Time frame: 2 years
Population: Per protocol. Analysis includes 289 malignant and 650 benign masses. One lesion was analyzed per participant.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion shape | 451 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion homogeneity-aggressive strategy | 510 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | median mean elasticity (Emean) | 407 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | diameter ration (Edia) | 333 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion homogeneity-conservative strategy | 464 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | median elasticity ratio (Erat) | 422 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | color score of Emax (Ecol)-conservative strategy | 457 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | maximum elasticity (Emax)-conservative strategy | 427 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | color score of Emax (Ecol)-aggressive strategy | 510 participants |
| Specificity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | maximum elasticity (Emax)-aggressive strategy | 503 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | color score of Emax (Ecol)-aggressive strategy | 285 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | median mean elasticity (Emean) | 285 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | median elasticity ratio (Erat) | 284 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion shape | 283 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion homogeneity-conservative strategy | 280 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | lesion homogeneity-aggressive strategy | 278 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | maximum elasticity (Emax)-conservative strategy | 286 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | maximum elasticity (Emax)-aggressive strategy | 281 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | diameter ration (Edia) | 286 participants |
| Sensitivity | Estimates of Effect of Selectively Upgrading BIRADS Category 3 and Downgrading BIRADS 4a Masses Based on SWE Features. Overall Specificity and Sensitivity of BI-RADS Score Using Conventional B-mode Ultrasound vs. B-mode + Certain SWE Characteristics | color score of Emax (Ecol)-conservative strategy | 288 participants |
Comparison: Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image specificity of the homogeneity of elasticity with the lesion and surrounding tissue using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if they were not homogeneous on the color overlay SW elastographic image; BIRADS 4a mass downgraded to BIRADS 3' if they were very homogeneous).~Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image specificity of the homogeneity of elasticity with the lesion and surrounding tissue using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if they were not homogeneous on the color overlay SW elastographic image; BIRADS 4a mass downgraded to BIRADS 3' if they were very or reasonably homogeneous).~Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image specificity of the maximum elasticity using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if maximum elasticity was 160 kiloPascals (kPa) (7.3 m/sec) or more; BIRADS 4a mass downgraded to BIRADS 3' if the maximum elasticity was 30 kPa (3.2 m/sec) or less).~Null hypothesis is that there is no difference between the two tests.p-value: 0.001McNemar
Comparison: Elastography image specificity of the maximum elasticity using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if maximum elasticity was 160 kPa(7.3 m/sec) or more; BIRADS 4a mass downgraded to BIRADS 3' if the maximum elasticity was 80 kPa (5.2 m/sec) or less).~Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image median mean elasticity value.~Null hypothesis is that there is no difference between the two tests.p-value: 0.18McNemar
Comparison: Elastography image diameter ratio.~Null hypothesis is that there is no difference between the two tests.p-value: 0.006McNemar
Comparison: Elastography image median elasticity ratio.~Null hypothesis is that there is no difference between the two tests.p-value: 0.006McNemar
Comparison: Elastography image specificity of color using a six-point color scale of maximum elasticity in the mass and surrounding parenchyma using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if the max stiffness on the color overlay SW elastographic image was red; BIRADS 4a mass downgraded to BIRADS 3' if the max stiffness on the color overlay SW elastographic image was black to dark blue).~Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image specificity of color using a six-point color scale of maximum elasticity in the mass and surrounding parenchyma using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if the max stiffness on the color overlay SW elastographic image was red; BIRADS 4a mass downgraded to BIRADS 3' if the max stiffness on the color overlay SW elastographic image was black to dark blue, or light blue).~Null hypothesis is that there is no difference between the two tests.p-value: <0.001McNemar
Comparison: Elastography image lesion shape.~Null hypothesis is that there is no difference between the two tests.p-value: 0.48McNemar
Comparison: Elastography image sensitivity of the homogeneity of elasticity with the lesion and surrounding tissue using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if they were not homogeneous on the color overlay SW elastographic image; BIRADS 4a mass downgraded to BIRADS 3' if they were very homogeneous).~Null hypothesis is that there is no difference between the two tests.p-value: 0.74McNemar
Comparison: Elastography image sensitivity of the homogeneity of elasticity with the lesion and surrounding tissue using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if they were not homogeneous on the color overlay SW elastographic image; BIRADS 4a mass downgraded to BIRADS 3' if they were very or reasonably homogeneous).~Null hypothesis is that there is no difference between the two tests.p-value: 0.37McNemar
Comparison: Elastography image sensitivity of the maximum elasticity using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if maximum elasticity was 160 kPa(7.3 m/sec) or more; BIRADS 4a mass downgraded to BIRADS 3' if the maximum elasticity was 30 kPa (3.2 m/sec) or less).~Null hypothesis is that there is no difference between the two tests.p-value: 0.025McNemar
Comparison: Elastography image sensitivity of the maximum elasticity using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if maximum elasticity was 160 kPa(7.3 m/sec) or more; BIRADS 4a mass downgraded to BIRADS 3' if the maximum elasticity was 80 kPa (5.2 m/sec) or less).~Null hypothesis is that there is no difference between the two tests.p-value: >0.99McNemar
Comparison: Elastography image median mean elasticity value.~Null hypothesis is that there is no difference between the two tests.p-value: 0.046McNemar
Comparison: Elastography image diameter ratio.~Null hypothesis is that there is no difference between the two tests.p-value: 0.025McNemar
Comparison: Elastography image median elasticity ratio.~Null hypothesis is that there is no difference between the two tests.p-value: 0.083McNemar
Comparison: Elastography image sensitivity of color using a six-point color scale of maximum elasticity in the mass and surrounding parenchyma using conservative strategy (BIRADS 3 mass upgraded to BIRADS 4a' if the max stiffness on the color overlay SW elastographic image was red; BIRADS 4a mass downgraded to BIRADS 3' if the max stiffness on the color overlay SW elastographic image was black to dark blue).~Null hypothesis is that there is no difference between the two tests.p-value: 0.008McNemar
Comparison: Elastography image sensitivity of color using a six-point color scale of maximum elasticity in the mass and surrounding parenchyma using aggressive strategy (BIRADS 3 mass upgraded to BIRADS 4a' if the max stiffness on the color overlay SW elastographic image was red; BIRADS 4a mass downgraded to BIRADS 3' if the max stiffness on the color overlay SW elastographic image was black to dark blue, or light blue).~Null hypothesis is that there is no difference between the two tests.p-value: 0.21McNemar
Secondary
Interobserver Agreement of B Mode Ultrasound and SWE Features
614 benign and 144 malignant breast masses.
Time frame: performed on the same day, after study completion
Population: Per protocol. One lesion was analyzed per participant. Two independent readers did a qualitative assessment of 7 image parameters (per lesion) using Kappa.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | B-mode margin | 0.38 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | B-mode BI-RADS assessment | 0.59 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | SWE shape | 0.40 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | SWE homogeneity | 0.57 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | SWE Emax | 0.66 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | B-mode shape | 0.58 kappa statistic |
| Specificity | Interobserver Agreement of B Mode Ultrasound and SWE Features | B-mode orientation | 0.53 kappa statistic |
Secondary
Intraobserver Reliability of Quantitative SWE Measurements
614 benign and 144 malignant lesions. Each category's measurement (diameter, area, etc..) is performed 3 times. These 3 measurements are then compared with each other in order to calculate the interclass correlation coefficient.
Time frame: performed on the same day, within 2 years from study start date
Population: Per protocol. 758 masses (614 benign and 144 malignant)are analyzed. One lesion was analyzed per participant. 3 acquisitions per lesion were analyzed with the ICC method for 10 different parameters.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | Diameter | 0.94 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | Area | 0.95 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E ratio (mass:fat) | 0.77 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E minumum fat | 0.57 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E maximum fat | 0.62 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E mean fat | 0.71 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | Perimeter | 0.95 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E minimum mass | 0.71 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E maximum mass | 0.84 correlation coefficient |
| Specificity | Intraobserver Reliability of Quantitative SWE Measurements | E mean mass | 0.87 correlation coefficient |
Secondary
Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature
614 benign and 144 malignant breast lesions were scanned in SWE 3 consecutive times, and the similarity of the 3 images was evaluated by the investigator.
Time frame: performed on the same day, within 2 years from study start date
Population: Per protocol. One lesion was analyzed per participant.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Specificity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | All images very or reasonably similar | 385 participants |
| Specificity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Images very dissimilar | 1 participants |
| Specificity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Some images similar | 5 participants |
| Sensitivity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | All images very or reasonably similar | 176 participants |
| Sensitivity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Some images similar | 34 participants |
| Sensitivity | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Images very dissimilar | 2 participants |
| Not Homogeneous | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | All images very or reasonably similar | 105 participants |
| Not Homogeneous | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Images very dissimilar | 5 participants |
| Not Homogeneous | Qualitative Intraobserver Reproducibility of SWE Related to Homogeneity Feature | Some images similar | 45 participants |
Comparison: Null hypothesis: no variance of similarity exists between the three groupsp-value: <0.001Kruskal-Wallis