Andrew Evans
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Andrew Evans has a total of 218 co-authors. Here are the top co-authors by number of shared publications:
- Patsy Whelehan
East of Scotland Breast Screening Programme, NHS Tayside, Dundee, UK
14 shared publications - Sarah Vinnicombe
Division of Cancer Research, Ninewells Hospital & Medical School, Dundee, UK
13 shared publications - Colin A. Purdie
Ninewells Hospital and Medical School
10 shared publications Alastair Thompson
Baylor College of Medicine, 1 Baylor Plaza, Houston, TX, 77030, United States; Population Health and Genomics, School of Medicine, University of Dundee, Ninewells Hospital and Medical School, Dundee, DD1 9SY, UK
9 shared publicationsLee Jordan
Breast Unit, Ninewells Hospital, Dundee, USA
6 shared publications- Ian Ogilvie Ellis
Department of Histopathology, Nottingham University Hospitals, Nottingham, UK
4 shared publications
Latest Publications
- Open Access
- 6 March 2024
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 97 (1157) , 882-885
Mammographic surveillance after breast cancer
Early detection of local recurrence has been shown to improve survival. What is unclear is how frequently mammography should be performed, how long surveillance should continue and how the answers to these questions vary with tumour pathology, patients age, and surgery type. Many of these questions are not directly answerable from the current literature. While some of these questions will be answered by the Mammo-50 study, evidence from local recurrence rates, tumour biology, and the lead time of mammography can be used to guide policy. Young age is the strongest predictor of local recurrence and given the short lead time of screening in women under 50, these women require annual mammography. Women over 50 with HER-2 positive and triple negative breast cancer have higher rates of local recurrence after breast conserving surgery than women with luminal cancers. Women with HER-2 positive and triple negative breast cancer also have a higher rate of recurrence in years 1-3 post surgery. Annual mammography in year 1-4 would appear justified. Women over 50 with luminal cancers have low rates of local recurrence and no early peak. Recurrence growth will be low due to tumour biology and hormone therapy. Biennial mammography after year 2 would seem appropriate. Women over 50 following mastectomy have no early peak in contralateral cancers so the frequency should be determined by the lead time of screening. This would suggest 2 yearly mammography for women aged 50-60 while 3 yearly mammography may suffice for women over 60.
- 6 March 2024
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 97 (1157) , 882-885
Open AccessMammographic surveillance after breast cancer
Early detection of local recurrence has been shown to improve survival. What is unclear is how frequently mammography should be performed, how long surveillance should continue and how the answers to these questions vary with tumour pathology, patients age, and surgery type. Many of these questions are not directly answerable from the current literature. While some of these questions will be answered by the Mammo-50 study, evidence from local recurrence rates, tumour biology, and the lead time of mammography can be used to guide policy. Young age is the strongest predictor of local recurrence and given the short lead time of screening in women under 50, these women require annual mammography. Women over 50 with HER-2 positive and triple negative breast cancer have higher rates of local recurrence after breast conserving surgery than women with luminal cancers. Women with HER-2 positive and triple negative breast cancer also have a higher rate of recurrence in years 1-3 post surgery. Annual mammography in year 1-4 would appear justified. Women over 50 with luminal cancers have low rates of local recurrence and no early peak. Recurrence growth will be low due to tumour biology and hormone therapy. Biennial mammography after year 2 would seem appropriate. Women over 50 following mastectomy have no early peak in contralateral cancers so the frequency should be determined by the lead time of screening. This would suggest 2 yearly mammography for women aged 50-60 while 3 yearly mammography may suffice for women over 60.
- Open Access
- 1 August 2023
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 96 (1148) , 20220921
Predicting the response to neoadjuvant chemotherapy. Can the addition of tomosynthesis improve the accuracy of contrast-enhanced spectral mammography? A comparison with breast MRI
Objectives: Image monitoring is essential to monitor response to neoadjuvant chemotherapy (NACT). Whilst breast MRI is the gold-standard technique, evidence suggests contrast-enhanced spectral mammography (CESM) is comparable. We investigate whether the addition of digital breast tomosynthesis (DBT) to CESM increases the accuracy of response prediction. Methods: Women receiving NACT for breast cancer were included. Imaging with CESM+DBT and MRI was performed post-NACT. Imaging appearance was compared with pathological specimens. Accuracy for predicting pathological complete response (pCR) and concordance with size of residual disease was calculated. Results: Sixteen cancers in 14 patients were included, 10 demonstrated pCR. Greatest accuracy for predicting pCR was with CESM enhancement (accuracy: 81.3%, sensitivity: 100%, specificity: 57.1%), followed by MRI (accuracy: 62.5%, sensitivity: 44.4%, specificity: 85.7%). Concordance with invasive tumour size was greater for CESM enhancement than MRI, concordance-coefficients 0.70 vs 0.66 respectively. MRI demonstrated greatest concordance with whole tumour size followed by CESM+microcalcification, concordance coefficients 0.86 vs 0.69. DBT did not improve accuracy for prediction of pCR or residual disease size. CESM+DBT underestimated size of residual disease, MRI overestimated but no significant differences were seen (p>0.05). Conclusions: CESM is similar to MRI for predicting residual disease post-NACT. Size of enhancement alone demonstrates best concordance with invasive disease. Inclusion of residual microcalcification improves concordance with ductal carcinoma in situ. The addition of DBT to CESM does not improve accuracy. Advances in knowledge: The addition ofDBT to CESM does not improve NACT response prediction. CESM enhancement has greatest accuracy for residual invasive disease, CESM+calcification has greater accuracy for residual in situ disease.
- 1 August 2023
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 96 (1148) , 20220921
Open AccessPredicting the response to neoadjuvant chemotherapy. Can the addition of tomosynthesis improve the accuracy of contrast-enhanced spectral mammography? A comparison with breast MRI
Objectives: Image monitoring is essential to monitor response to neoadjuvant chemotherapy (NACT). Whilst breast MRI is the gold-standard technique, evidence suggests contrast-enhanced spectral mammography (CESM) is comparable. We investigate whether the addition of digital breast tomosynthesis (DBT) to CESM increases the accuracy of response prediction. Methods: Women receiving NACT for breast cancer were included. Imaging with CESM+DBT and MRI was performed post-NACT. Imaging appearance was compared with pathological specimens. Accuracy for predicting pathological complete response (pCR) and concordance with size of residual disease was calculated. Results: Sixteen cancers in 14 patients were included, 10 demonstrated pCR. Greatest accuracy for predicting pCR was with CESM enhancement (accuracy: 81.3%, sensitivity: 100%, specificity: 57.1%), followed by MRI (accuracy: 62.5%, sensitivity: 44.4%, specificity: 85.7%). Concordance with invasive tumour size was greater for CESM enhancement than MRI, concordance-coefficients 0.70 vs 0.66 respectively. MRI demonstrated greatest concordance with whole tumour size followed by CESM+microcalcification, concordance coefficients 0.86 vs 0.69. DBT did not improve accuracy for prediction of pCR or residual disease size. CESM+DBT underestimated size of residual disease, MRI overestimated but no significant differences were seen (p>0.05). Conclusions: CESM is similar to MRI for predicting residual disease post-NACT. Size of enhancement alone demonstrates best concordance with invasive disease. Inclusion of residual microcalcification improves concordance with ductal carcinoma in situ. The addition of DBT to CESM does not improve accuracy. Advances in knowledge: The addition ofDBT to CESM does not improve NACT response prediction. CESM enhancement has greatest accuracy for residual invasive disease, CESM+calcification has greater accuracy for residual in situ disease.
- Open Access
- 1 February 2023
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 96 (1142) , 20220078
Fibroepithelial lesions of the breast: improving the accuracy of imaging diagnosis and reducing unnecessary biopsy
Fibroadenomas and phyllodes tumours are fibroepithelial lesions of the breast. Fibroadenomas are common benign breast masses encountered both symptomatically and incidentally. Phyllodes tumours are uncommon and usually present symptomatically. Management of fibroadenomas focuses on reducing biopsies without missing cancers while radiological management of phyllodes tumours should focus on predicting the need for surgical excision with or without margins.
- 1 February 2023
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 96 (1142) , 20220078
Open AccessFibroepithelial lesions of the breast: improving the accuracy of imaging diagnosis and reducing unnecessary biopsy
Fibroadenomas and phyllodes tumours are fibroepithelial lesions of the breast. Fibroadenomas are common benign breast masses encountered both symptomatically and incidentally. Phyllodes tumours are uncommon and usually present symptomatically. Management of fibroadenomas focuses on reducing biopsies without missing cancers while radiological management of phyllodes tumours should focus on predicting the need for surgical excision with or without margins.
- 1 November 2022
- Published by Springer Nature
- p. 113-126
Breast Ultrasound: Advanced Techniques
- Open Access
- 28 March 2022
- Published by Copernicus GmbH
Improving resilience through a surface water flooding decision support system
Flood Foresight is JBA’s strategic flood monitoring and forecasting system, providing flood inundation and depth estimates across the UK and Ireland at 30m resolution up to 10-days ahead of fluvial flood events. It consists of Flood Monitoring (based on observed discharges from river gauge telemetry) and Flood Forecasting (based on simulated discharge from a rainfall-runoff model) modules.Recently, Flood Foresight has been expanded to provide asset alerting around heavy rainfall and surface water (pluvial) flooding, demonstrated in a proof-of-concept system on behalf of Network Rail during a Small Business Research Initiative project funded by Department for Transport and delivered by InnovateUK.The surface water flood forecasting system is now running in real time using high resolution ensemble rainfall forecasts from Met Eireann (IREPS). This system represents a major advance in the availability of information indicating the risk to rail infrastructure across Great Britain. Taking advantage of ensemble rainfall forecasts, it is possible to give an indication of where rain might happen and the severity of that rain (in comparison to historical rainfall amounts), and also to provide an indication of the confidence in that forecast. This concept is crucial to the handling of intense rainfall events, due to their inherent lack of predictability. The presentation of mapped likelihood information for both rainfall and surface water flooding forecasts provides users with spatial context for the asset alerts. It allows them to see the extent and uncertainty in the location of the intense rainfall event. The system has been developed to run autonomously using rainfall forecasts as they are provided by Met Eireann, via FTP. Therefore the resulting asset alert information is always available, and always presents the most up-to-date information. This gives asset managers the ability to access the information at a time that is convenient to them, but also the system can provide alerts when assets are identified as at risk as the information becomes available. The forecast data is available beyond 36 hours into the future, providing sufficient lead time for asset managers to coordinate responses and mobilise staff and equipment, if needed. The temporal resolution of the forecast information is high at short lead times (i.e. hourly for the first 6 hours), decreasing as lead time increases (after 24 hours the information is 6 hourly, further reducing to 12 hourly when longer lead time forecasts are available). This decreasing temporal resolution with longer lead times allows for increased uncertainty in the timing of events further in the future to be obscured to the user, reducing confusion if the timing changes with subsequent forecasts. The proof-of-concept system focuses on the rail industry, however it is extensible to other sectors where population, assets or infrastructure are vulnerable to surface water flooding. Flood impact data and associated alerts can be customised based on a client’s asset portfolio and their incident management needs.The presentation will explore heavy rainfall events evaluated during the proof-of-concept demonstrations, describing the information the Flood Foresight system could have provided ahead of, and during the event.
- 28 March 2022
- Published by Copernicus GmbH
Open AccessImproving resilience through a surface water flooding decision support system
Flood Foresight is JBA’s strategic flood monitoring and forecasting system, providing flood inundation and depth estimates across the UK and Ireland at 30m resolution up to 10-days ahead of fluvial flood events. It consists of Flood Monitoring (based on observed discharges from river gauge telemetry) and Flood Forecasting (based on simulated discharge from a rainfall-runoff model) modules.Recently, Flood Foresight has been expanded to provide asset alerting around heavy rainfall and surface water (pluvial) flooding, demonstrated in a proof-of-concept system on behalf of Network Rail during a Small Business Research Initiative project funded by Department for Transport and delivered by InnovateUK.The surface water flood forecasting system is now running in real time using high resolution ensemble rainfall forecasts from Met Eireann (IREPS). This system represents a major advance in the availability of information indicating the risk to rail infrastructure across Great Britain. Taking advantage of ensemble rainfall forecasts, it is possible to give an indication of where rain might happen and the severity of that rain (in comparison to historical rainfall amounts), and also to provide an indication of the confidence in that forecast. This concept is crucial to the handling of intense rainfall events, due to their inherent lack of predictability. The presentation of mapped likelihood information for both rainfall and surface water flooding forecasts provides users with spatial context for the asset alerts. It allows them to see the extent and uncertainty in the location of the intense rainfall event. The system has been developed to run autonomously using rainfall forecasts as they are provided by Met Eireann, via FTP. Therefore the resulting asset alert information is always available, and always presents the most up-to-date information. This gives asset managers the ability to access the information at a time that is convenient to them, but also the system can provide alerts when assets are identified as at risk as the information becomes available. The forecast data is available beyond 36 hours into the future, providing sufficient lead time for asset managers to coordinate responses and mobilise staff and equipment, if needed. The temporal resolution of the forecast information is high at short lead times (i.e. hourly for the first 6 hours), decreasing as lead time increases (after 24 hours the information is 6 hourly, further reducing to 12 hourly when longer lead time forecasts are available). This decreasing temporal resolution with longer lead times allows for increased uncertainty in the timing of events further in the future to be obscured to the user, reducing confusion if the timing changes with subsequent forecasts. The proof-of-concept system focuses on the rail industry, however it is extensible to other sectors where population, assets or infrastructure are vulnerable to surface water flooding. Flood impact data and associated alerts can be customised based on a client’s asset portfolio and their incident management needs.The presentation will explore heavy rainfall events evaluated during the proof-of-concept demonstrations, describing the information the Flood Foresight system could have provided ahead of, and during the event.
- Open Access
- 18 February 2022
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 95 (1134) , 20210779
Contrast-enhanced digital breast tomosythesis and breast MRI to monitor response to neoadjuvant chemotherapy: patient tolerance and preference
Objective: Contrast-enhanced digital breast tomosynthesis (CE-DBT) is a novel imaging technique, combining contrast-enhanced spectral mammography and tomosynthesis. This may offer an alternative imaging technique to breast MRI for monitoring of response to neoadjuvant chemotherapy. This paper addresses patient experience and preference regarding the two techniques. Methods: Conducted as part of a prospective pilot study; patients were asked to complete questionnaires pertaining to their experience of CE-DBT and MRI following pre-treatment and end-of-treatment imaging. Questionnaires consisted of eight questions answered on a categorical scale, two using a visual analogue scale (VAS), and a question to indicate preference of imaging technique. Statistical analysis was performed with Wilcoxon signed rank test and McNemar test for related samples using SPSS v. 25. Results: 18 patients were enrolled in the pilot study. Matched CE-DBT and MRI questionnaires were completed after 22 patient episodes. Patient preference was indicated after 31 patient episodes. Overall, on 77% of occasions patients preferred CE-DBT with no difference between pre-treatment and end-of-treatment imaging. Overall experience (p = 0.008), non-breast pain (p = 0.046), anxiety measured using VAS (p = 0.003), and feeling of being put at ease by staff (p = 0.023) was better for CE-DBT. However, more breast pain was experienced during CE-DBT when measured on both VAS (p = 0.011) and categorical scale (p = 0.021). Conclusion: Our paper suggests that patients prefer CE-DBT to MRI, adding further evidence in favour of contrast-enhanced mammographic techniques. Advances in knowledge: Contrast mammographic techniques offer an alternative, more accessible imaging technique to breast MRI. Whilst other studies have addressed patient experience of contrast-enhanced spectral mammography, this is the first study to directly explore patient preference for CE-DBT over MRI in the setting of neoadjuvant chemotherapy, finding that overall, patients preferred CE-DBT despite the relatively long breast compression.
- 18 February 2022
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 95 (1134) , 20210779
Open AccessContrast-enhanced digital breast tomosythesis and breast MRI to monitor response to neoadjuvant chemotherapy: patient tolerance and preference
Objective: Contrast-enhanced digital breast tomosynthesis (CE-DBT) is a novel imaging technique, combining contrast-enhanced spectral mammography and tomosynthesis. This may offer an alternative imaging technique to breast MRI for monitoring of response to neoadjuvant chemotherapy. This paper addresses patient experience and preference regarding the two techniques. Methods: Conducted as part of a prospective pilot study; patients were asked to complete questionnaires pertaining to their experience of CE-DBT and MRI following pre-treatment and end-of-treatment imaging. Questionnaires consisted of eight questions answered on a categorical scale, two using a visual analogue scale (VAS), and a question to indicate preference of imaging technique. Statistical analysis was performed with Wilcoxon signed rank test and McNemar test for related samples using SPSS v. 25. Results: 18 patients were enrolled in the pilot study. Matched CE-DBT and MRI questionnaires were completed after 22 patient episodes. Patient preference was indicated after 31 patient episodes. Overall, on 77% of occasions patients preferred CE-DBT with no difference between pre-treatment and end-of-treatment imaging. Overall experience (p = 0.008), non-breast pain (p = 0.046), anxiety measured using VAS (p = 0.003), and feeling of being put at ease by staff (p = 0.023) was better for CE-DBT. However, more breast pain was experienced during CE-DBT when measured on both VAS (p = 0.011) and categorical scale (p = 0.021). Conclusion: Our paper suggests that patients prefer CE-DBT to MRI, adding further evidence in favour of contrast-enhanced mammographic techniques. Advances in knowledge: Contrast mammographic techniques offer an alternative, more accessible imaging technique to breast MRI. Whilst other studies have addressed patient experience of contrast-enhanced spectral mammography, this is the first study to directly explore patient preference for CE-DBT over MRI in the setting of neoadjuvant chemotherapy, finding that overall, patients preferred CE-DBT despite the relatively long breast compression.
- Open Access
- 3 December 2021
- Published by Elsevier in The Breast
- Vol. 61, 11-21
A pre-operative prognostic model predicting all cause and cause specific mortality for women presenting with invasive breast cancer
- 3 December 2021
- Published by Elsevier in The Breast
- Vol. 61, 11-21
Open AccessA pre-operative prognostic model predicting all cause and cause specific mortality for women presenting with invasive breast cancer
- 1 December 2021
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 94 (1128)
A retrospective review of MRI features associated with metastasis-free survival in women with breast cancer: focusing on skin thickening and skin enhancement
Objectives: To identify associations between MRI-detected skin thickening and enhancement and metastasis-free survival (MFS) given recent reports of skin thickening on ultrasound being a poorer prognostic indicator. Methods: Interrogation of a prospectively collected database of ultrasound-visible breast lesions showed 214 lesions with pre-treatment MRIs available for analysis in a single centre. Data on MFS was prospectively collected. Retrospective MRI review was performed blinded to outcome. Imaging factors recorded were presence of skin thickening and enhancement, non-mass-enhancement (NME) and abnormal nodes, mass characteristics, perilesional oedema and background parenchymal enhancement. Statistical analysis used chi-squared test, Kaplan-Meier survival curves, the Log-rank test and receiver-operator characteristic (ROC) curves. Results: During a median follow-up period of 5.6 years, 21 (10%) of 212 patients developed distant metastases. Skin thickening [24 of 30 (80%) vs 169 of 184 (92%), p = 0.043] and skin enhancement [15 of 20 (75%) vs 178 of 194 (92%), p = 0.016] were associated with poorer MFS. Large index lesion size [p < 0.001, AUC 0.823], large sum of masses [p < 0.001, AUC 0.813], increasing total lesion extent including NME [p < 0.001, AUC 0.749] and abnormal axillary nodes [55 of 66 (83%) vs 138 of 148 (93%), p = 0.024] were also associated with poorer MFS. Conclusions: Skin thickening and enhancement on breast MRI are associated with poorer MFS. These findings should be taken into account when managing patients with invasive breast cancer. Advances in knowledge: Skin enhancement on breast MRI at diagnosis is associated with metastases development. Skin thickening on breast MRI is associated with future metastatic disease.
- 1 December 2021
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 94 (1128)
A retrospective review of MRI features associated with metastasis-free survival in women with breast cancer: focusing on skin thickening and skin enhancement
Objectives: To identify associations between MRI-detected skin thickening and enhancement and metastasis-free survival (MFS) given recent reports of skin thickening on ultrasound being a poorer prognostic indicator. Methods: Interrogation of a prospectively collected database of ultrasound-visible breast lesions showed 214 lesions with pre-treatment MRIs available for analysis in a single centre. Data on MFS was prospectively collected. Retrospective MRI review was performed blinded to outcome. Imaging factors recorded were presence of skin thickening and enhancement, non-mass-enhancement (NME) and abnormal nodes, mass characteristics, perilesional oedema and background parenchymal enhancement. Statistical analysis used chi-squared test, Kaplan-Meier survival curves, the Log-rank test and receiver-operator characteristic (ROC) curves. Results: During a median follow-up period of 5.6 years, 21 (10%) of 212 patients developed distant metastases. Skin thickening [24 of 30 (80%) vs 169 of 184 (92%), p = 0.043] and skin enhancement [15 of 20 (75%) vs 178 of 194 (92%), p = 0.016] were associated with poorer MFS. Large index lesion size [p < 0.001, AUC 0.823], large sum of masses [p < 0.001, AUC 0.813], increasing total lesion extent including NME [p < 0.001, AUC 0.749] and abnormal axillary nodes [55 of 66 (83%) vs 138 of 148 (93%), p = 0.024] were also associated with poorer MFS. Conclusions: Skin thickening and enhancement on breast MRI are associated with poorer MFS. These findings should be taken into account when managing patients with invasive breast cancer. Advances in knowledge: Skin enhancement on breast MRI at diagnosis is associated with metastases development. Skin thickening on breast MRI is associated with future metastatic disease.
- Open Access
- 15 November 2021
- Published by Springer Nature in Breast Cancer
- Vol. 29 (2) , 296-301
The value of prognostic ultrasound features of breast cancer in different molecular subtypes with a focus on triple negative disease
The ultrasound (US) features of breast cancer have recently been shown to have prognostic significance. We aim to assess these features according to molecular subtype. 1140 consecutive US visible invasive breast cancers had US size and mean stiffness by shearwave elastography (SWE) recorded prospectively. Skin thickening (> 2.5 mm) overlying the cancer on US and the presence of posterior echo enhancement were assessed retrospectively while blinded to outcomes. Cancers were classified as luminal, triple negative (TN) or HER2 + ve based on immunohistochemistry and florescent in-situ hybridization. The relationship between US parameters and breast cancer specific survival (BCSS) was ascertained using Kaplan–Meier survival curves and ROC analysis. At median follow-up 6.3 year, there were 117 breast cancer (10%) and 132 non-breast deaths (12%). US size was significantly associated with BCSS all groups (area under the curve (AUC) 0.74 in luminal cancers, 0.64 for TN and 0.65 for HER2 + ve cancers). US skin thickening was associated most strongly with poor prognosis in TN cancers (53% vs. 80% 6 year survival, p = 0.0004). Posterior echo enhancement was associated with a poor BCSS in TN cancers (63% vs. 82% 6 year survival, p = 0.02). Mean stiffness at SWE was prognostic in the luminal and HER2 positive groups (AUC 0.69 and 0.63, respectively). In the subgroup of patients with TN cancers receiving neo-adjuvant chemotherapy posterior enhancement and skin thickening were not associated with response. US skin thickening is a poor prognostic indicator is all 3 subtypes studied, while posterior enhancement was associated with poor outcome in TN cancers
- 15 November 2021
- Published by Springer Nature in Breast Cancer
- Vol. 29 (2) , 296-301
Open AccessThe value of prognostic ultrasound features of breast cancer in different molecular subtypes with a focus on triple negative disease
The ultrasound (US) features of breast cancer have recently been shown to have prognostic significance. We aim to assess these features according to molecular subtype. 1140 consecutive US visible invasive breast cancers had US size and mean stiffness by shearwave elastography (SWE) recorded prospectively. Skin thickening (> 2.5 mm) overlying the cancer on US and the presence of posterior echo enhancement were assessed retrospectively while blinded to outcomes. Cancers were classified as luminal, triple negative (TN) or HER2 + ve based on immunohistochemistry and florescent in-situ hybridization. The relationship between US parameters and breast cancer specific survival (BCSS) was ascertained using Kaplan–Meier survival curves and ROC analysis. At median follow-up 6.3 year, there were 117 breast cancer (10%) and 132 non-breast deaths (12%). US size was significantly associated with BCSS all groups (area under the curve (AUC) 0.74 in luminal cancers, 0.64 for TN and 0.65 for HER2 + ve cancers). US skin thickening was associated most strongly with poor prognosis in TN cancers (53% vs. 80% 6 year survival, p = 0.0004). Posterior echo enhancement was associated with a poor BCSS in TN cancers (63% vs. 82% 6 year survival, p = 0.02). Mean stiffness at SWE was prognostic in the luminal and HER2 positive groups (AUC 0.69 and 0.63, respectively). In the subgroup of patients with TN cancers receiving neo-adjuvant chemotherapy posterior enhancement and skin thickening were not associated with response. US skin thickening is a poor prognostic indicator is all 3 subtypes studied, while posterior enhancement was associated with poor outcome in TN cancers