RPS 2202 - Advanced breast imaging and interventions: CT and minimally invasive approaches

Author Block: E. D’Ascoli, C. Depretto, G. Rossini, G. Irmici, C. De Berardinis, G. Della Pepa, A. Bonanomi, L. Corradini, G. P. Scaperrotta; Milan/IT
Purpose: To evaluate the role of US-guided Vacuum-assisted excision (VAE) in the management of imaging-histological discordant lesions after core needle biopsy (CNB), avoiding unnecessary surgical excision.
Methods or Background: We retrospectively evaluated all patients who underwent US-guided VAE with a 9 G needle at our Institution between January 2015 and August 2023 for imaging-histological discordant lesions after CNB with a 14 G needle. We determined the upgrade rate to malignancy (UR) after VAE and delayed false-negative results after surgery or follow-up. We also assessed the diagnostic performance of the procedure, technical efficacy (TE), technical success (TS), the occurrence of complications and the correlation between upgrade to malignancy and lesion size, BI-RADS category, morphological characteristics, presence of calcifications and history of breast cancer (χ² and Mann-Whitney tests).
Results or Findings: We collected data from 334 patients. After VAE, 65.3% (N=218) of the excised lesions were classified as benign, 25% (N=85) were B3 lesions and 9.3% (N=31) were malignant (B5). The upgrade rate to malignancy after VAE was 9.3%. The delayed false negative results were 1.3%. Accuracy was 98.2%, SE was 88.6%, SP was 100%, PPV was 100% and NPV was 97.9%. TE of the procedure was 84.4%; TS was 100%. No major complications were observed. BIRADS category, morphology, margins, and orientation of the lesion were significantly associated with upgrade to malignancy (p<0.001).
Conclusion: VAE reduces the risk of underestimation of malignancy in cases of imaging-histological discordance, thus avoiding the need for surgery and reducing patients’ stress, overtreatment, complications and costs, without compromising the quality of patient care.
Limitations: Retrospective monocentric study with a limited number of patients.
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Independent Ethics Committee at the Fondazione IRCCS Istituto Nazionale dei Tumori, Milano.

Author Block: W. Sanderink¹, C. Meeuwis², S. De Waard³, A. Zouzos³, E. E. J. M. Wolters-Van Der Ben⁴, F. Van Vliet - Moret⁴, R. F. Lim⁵, J. Camps Herrero⁶, R. Mann¹; ¹Nijmegen/NL, ²Arnhem/NL, ³Stockholm/SE, ⁴Nieuwegein/NL, ⁵Bergen Op Zoom/NL, ⁶Alzira/ES
Purpose: To assess feasibility of complete removal of small breast cancers using Vacuum Assisted Biopsy (VAB) as a therapeutic procedure under ultrasound (US) guidance.
Methods or Background: From August 2021 to May 2025, 35 patients with invasive carcinomas diagnosed by core needle biopsy (CNB), measuring ≤15 mm on US and MRI were enrolled in this international, multicenter, prospective study. One patient withdrew after VAB and was excluded from analysis. All lesions were excised under local anesthesia using a 7G VAB system under US guidance. Three weeks later, breast-conserving surgery excised the VAE cavity. Histopathological findings from VAB were compared with subsequent surgical specimens, and the rate of complete or marginally involved tumor excision by VAB was assessed.
Results or Findings: 34 patients completed the study (mean age, 63.3 years; 13 Netherlands, 20 Spain, 1 Sweden). The mean MRI lesion size was 10.1 mm (range 4-13.7 mm). VAB histopathology identified 33 invasive carcinomas of no special type and 1 with medullary features. Complete excision by VAB was achieved in 12 of 34 patients (35.3%). An additional 6 patients had marginally involved tumor excisions, meeting Dutch guidelines for successful removal (18/34, 52.9%). 13 patients (38.2%) showed more than focal margin involvement (mean residual tumor size, 9.4 mm; range, 5-13 mm). In 3 patients, residual tumor was present in the surgical specimen, though margin involvement >4 mm was unclear.
Conclusion: US-guided VAB achieved complete or marginally adequate excision in a subset of small breast cancers. Residual tumor in several cases indicates that careful patient selection is essential before considering VAB as a standalone treatment.
Limitations: The small sample size and the need for further analysis of patient and procedural variables limit the ability to identify factors that may predict successful VAB excision.
Funding for this study: The study was an investigator initiated study sponsored by an unrestricted grant from BD.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: METC-Oost Nederland

Author Block: F. Di Naro, S. E. Baldi Giorgi, F. Pugliese, D. De Benedetto, G. Migliaro, S. Vidali, C. Bellini, T. Amadori, J. Nori; Florence/IT
Purpose: To evaluate the most effective non-surgical treatment for breast cancer in surgery-ineligible patients, comparing ultrasound-guided Cryoablation combined with hormonal therapy (HT) versus Cryoablation and hormonal therapy alone.
Methods or Background: 111 patients (mean age 81.2 years) not-suitable for surgery due to comorbidities or advanced age was
enrolled, total of 125 biopsy-confirmed breast cancers. All the lesions were invasive ductal carcinomas (mean size 17.7 mm), hormone-positive and HER2-negative, with no ultrasound-visible lymph-node involvement. Of these, 41 treated with Cryoablation and HT, 36 with Cryoablation and 36 with HT . Locoregional staging was performed with contrast-enhanced-mammography (CEM), followed by CEM and ultrasound follow-up at 12 months post-treatment. Only patients completing follow-up were included. Lesion size was compared at baseline and 12 months after-treatment.
Results or Findings: Of the 125 lesions, 98 completed the 12-months follow-up; 39 in the Cryoablation+HT group, 23 in the Cryoablation-only group, and 36 in the HT-only group. Tumor size reduction was significantly different between the groups (P=0.0005), with greatest reduction in the Cryoablation-with-HT group (83.3%, mean reduction 13.6 mm), followed by Cryoablation-only (61.7%, mean reduction 8.2 mm), and HT-only (42.1%, mean reduction 7.4 mm).Tumors with no CEM-enhancement, suggesting no residual-disease, were most frequent in the Cryoablation-with-HT group (69.2%), followed by Cryoablation-only (60.9%),and HT-only (36.1%). Tumors with complete remission (CR, RECIST 1.1) were similar between the Cryoablation-with-HT and Cryoablation-only groups (74.4% and 78.3%, respectively), followed by the HT-only (36.1%). Pairwise comparisons revealed significant differences between Cryoablation-with- HT and HT-only groups for CEM-enhancement, size reduction, and CR, expressing the added value of Cryoablation (p=0.0041, P<0.0001 and P<0.0001,respectively).
Conclusion: Cryoablation with hormonal-therapy significantly reduces tumor size and residual disease more effectively than therapy alone, making it a promising option for patients not-eligible for surgery.
Limitations: Single-center-retrospective
No-long-term follow-up
Small-cohort
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Meyer hospital florence

Author Block: M. Pasculli¹, E. Giannotti²; ¹Rome/IT, ²Cambridge/UK
Purpose: Our primary aim was to compare infection rates between sterile and aseptic technique.
Secondary aims included identifying possible predictors of infection—such as hematoma, diabetes, or immunosuppression—and explore options to reduce environmental impact.
Methods or Background: This prospective multicenter study included 15 centers: 6 using sterile technique (sterile gloves, gel, probe cover, gauze) and 9 using aseptic technique (clean but not fully sterile). Consecutive patients undergoing percutaneous breast biopsy were enrolled over two months (≥50 per center). At 1-month follow-up, patients completed a questionnaire on biopsy type, infection (medical assistance, antibiotics), pain, hematoma, comorbidities, and lesion characteristics. Infection rates were compared between sterile vs aseptic groups and CNB vs VAB using Fisher’s exact test, with risk ratios and Wilson 95% CIs. Due to rare events, Firth’s penalized logistic regression was applied. Infections were defined as cases requiring antibiotics post-biopsy.
Results or Findings: We analyzed 1,560 biopsies (CNB 1,249; VAB 311), of which 531 were performed under sterile and 1,029 under aseptic conditions. Overall, 9 infections (0.58%) occurred. Infection rates were 0.94% (5/531) in the sterile group and 0.39% (4/1,029) in the aseptic group (risk ratio 2.42, 95% CI 0.65–8.98; p=0.18). No significant differences were observed. Hematoma occurred in 6.4% of CNB and 15.4% of VAB. No infections occurred among patients with diabetes (n=45) or under immunosuppressive therapy (n=18).
Conclusion: Infections after breast biopsy were extremely rare, occurring in less than 1% of procedures. We found no difference between sterile and aseptic technique, or between core biopsy and vacuum-assisted biopsy. The aseptic approach is safe, faster, cheaper, and environmentally friendly. Overall, our results support best practice guidelines that integrate patient safety with sustainability.
Limitations: The low number of infections limited statistical power and led to imprecise estimates in stratified analyses.
Funding for this study: No funding was required for this study
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethical approval was obtained in some centers, while in others it was not required, in accordance with their local institutional policies.

Author Block: R. Duteweert¹, M. Schyns¹, L. Deden¹, C. Meeuwis¹, M. Dorrius²; ¹Arnhem/NL, ²Groningen/NL
Purpose: In women with breast cancer and dense breasts, additional contrast-enhanced imaging modalities are needed to determine the extent of the disease. This study aimed to evaluate the feasibility of spectral CT for breast cancer detection in these women, compared to MRI , using quantitative spectral analysis and qualitative image assessment.
Methods or Background: This single-centre feasibility study prospectively included 12 patients with locoregionally advanced breast cancer and dense breasts (ACR 3–4) between February 2023 and May 2025. All patients underwent breast MRI and 18F-FDG PET/CT as part of routine staging, followed by an additional prone spectral CT on a dual-layer spectral CT scanner. Quantitative analyses included spectral slope (λHU, defined as the attenuation slope between 40 and 70 keV) and iodine concentration in 21 histopathologically confirmed breast lesions (20 malignant and 1 benign), fibroglandular tissue, and pectoral muscle. Three radiologists independently conducted the qualitative image assessment, which comprised evaluation of background enhancement and lesion detection, with outcomes compared to MRI.
Results or Findings: Spectral CT allowed quantitative identification of breast cancer in all patients. Lesions showed a mean λHU of 2.8±1.2, compared to −0.4±0.4 in fibroglandular tissue and 0.7±0.7 in pectoral muscle. Mean iodine concentration was 1.4±0.6mg/mL in lesions, higher than in pectoral muscle (0.4±0.3mg/mL) and fibroglandular tissue (0.0±0.1mg/mL). Background enhancement was scored lower on spectral CT than on MRI across all patients (p=0.07). For lesion detection, spectral CT yielded 56 true positives, 7 false negatives, and 3 false positives, distributed across readers, compared with 55, 8, and 0 on MRI, respectively.
Conclusion: Spectral CT is feasible for breast cancer detection in women with dense breasts, providing quantitative lesion characterization and reader performance comparable to MRI.
Limitations: No pathological proof for all reader-marked lesions, therefore excluded from analysis.
Funding for this study: No
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: METC Oost Nederland

Author Block: S. Joomun¹, M. Nicoara¹, L. J. Oostveen², I. Sechopoulos², G. De Bock¹, M. Greuter¹, M. Dorrius¹; ¹Groningen/NL, ²Nijmegen/NL
Purpose: To evaluate influence of dose, reconstruction kernel, matrix size, and virtual monoenergetic imaging (VMI) level on breast lesion detectability on conventional energy integrated (EID) computed tomography (CT) and photon-counting detector (PCD) CT.
Methods or Background: A semi-anthropomorphic breast phantom containing hyperdense masses, calcification clusters, and fibers of different diameters was attached to an anthropomorphic thorax phantom and positioned on a dedicated breast support cushion. Scans were performed in the prone position, with the breast hanging freely, on EID-CT (Somatom Force, Siemens) and PCD-CT (Naeotom Alpha, Siemens). Dose (CTDIvol) and reconstruction kernel were varied on both systems, while matrix size and VMI level were varied on the PCD-CT system. Qualitative analysis was assessed by lesion visibility using a 5-point Likert scale.
Results or Findings: 0.40mm calcification cluster was visible on both systems above 21mGy, whereas 0.29mm cluster was only visible on PCD-CT above 29mGy. 1.80mm mass was not detected on EID-CT but was visible on PCD-CT above 29mGy. On EID-CT, 3.18, 4.76, and 6.32mm masses were visible above 7.4, 1.8, and 0.9mGy, respectively, while on PCD-CT, the 3.18mm mass was detectable above 3.7mGy and the 4.76 and 6.32mm masses above 0.9mGy. Fibers of 0.41 and 0.60mm were visible on EID-CT above 7.4 and 1.8mGy, whereas on PCD-CT, fibers of 0.23, 0.41, and 0.60mm were detected above 29, 7.4, and 0.9mGy, respectively. Optimal visibility was achieved at VMI 70 keV. Sharper kernel enhanced visibility of smaller structures. Use of 1024 matrix on PCD-CT improved detection of smaller structures.
Conclusion: PCD-CT demonstrated higher detectability of breast lesions compared with EID-CT, particularly for small structures.
Limitations: The breast phantom used in this study only contained adipose tissue and no glandular tissue.
Funding for this study: Funding was received from KWF PPS and Siemens Healthineers
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: F. V. Amstel¹, M. Lenaerts¹, F. M. Mottaghy¹, J. Simons², M. Smidt¹, T. Van Nijnatten¹; ¹Maastricht/NL, ²Rotterdam/NL
Purpose: The diagnostic accuracy and prognostic relevance of 18F-FDG PET/CT for regional nodal response assessment after neoadjuvant chemo-immunotherapy (NAC) in clinically node-positive (cN+) breast cancer remains unclear. Therefore, this review and meta-analysis aims to determine the diagnostic performance of post-NAC 18F-FDG PET/CT for axillary lymph node (ALN) response assessment and to evaluate the prognostic value of post-NAC 18F-FDG PET/CT for internal mammary lymph node (IMLN), and periclavicular lymph node response assessment.
Methods or Background: PubMed, Embase, Cochrane, and Web of Science were searched for studies using post-NAC 18F-FDG PET/CT for nodal response assessment in cN+ patients treated with NAC, up to 17/02/2025. Pooled sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. For ALNs, surgical pathology served as reference standard. For IMLNs and periclavicular lymph nodes, locoregional or distant recurrence was used as surrogate reference standard.
Results or Findings: Eight studies (574 patients) were included: six (558 patients) on ALNs, two (16 patients) on IMLNs, and one (1 patient) on periclavicular lymph nodes. Regarding ALNs, the overall axillary pathologic complete response rate was 45.2%. The pooled sensitivity, specificity, PPV, and NPV were 52% (95% CI:0.24-0.79), 83% (95% CI:0.72-0.91), 78% (95% CI:0.62-0.89), and 59% (95% CI:0.30-0.82). Regarding IMLNs, the pooled sensitivity, specificity, PPV, and NPV were 40% (95% CI:0.05-0.85), 100% (95% CI:0.72-1.00), 100% (95% CI:0.16-1.00), and 78.6% (95% CI:0.49-0.95). Regarding periclavicular lymph nodes, findings were limited to only one patient.
Conclusion: The diagnostic performance of post-NAC 18F-FDG PET/CT to accurately identify ALN response is limited. Regarding post-NAC 18F-FDG PET/CT in assessing IMLN and periclavicular lymph node response, additional evidence is needed to determine prognostic relevance.
Limitations: For periclavicular lymph nodes, data pooling was not possible given low number of included studies
Funding for this study: Florien J.G. van Amstel received funding for this study by the Dutch Cancer Society (REFINE-trial; project 14055). Melissa Lenaerts received funding from Kankeronderzoeksfonds Limburg. None of the other authors received support from any organization for the submitted work.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: M. Elghobashy, P. Fletcher, I. Masroor, F. Peters, S-W. Tan; Stevenage/UK
Purpose: This study aims to demonstrate a centre's experience of the utility of the RCR guidelines for CT staging in breast cancer.
Methods or Background: In the United Kingdom, around 57,000 new cases of breast cancer are diagnosed annually [1]. Early identification of metastatic disease is crucial for improving survival and morbidity. The Royal College of Radiologists (RCR) recommends use of whole-body CT staging for distant metastatic disease in those presenting with T3/T4 cancer, those with 4 or more abnormal lymph nodes at axillary ultrasound and those with symptoms that raise suspicion of metastasis [2].

Despite clear guidelines, patients often undergo CT imaging without fulfilling the aforementioned criteria, thus leading to increased and potentially unnecessary radiation exposure. This study explores the utility of the RCR guidelines in stratifying patients with and without metastasis.
Results or Findings: A retrospective observational study was undertaken with data extracted from a 5-year period. 526 patients were identified who had a breast cancer staging CT. 214 of these studies were excluded from analysis as they represented patients with disease recurrence, known metastatic disease or treatment surveillance. Of the remaining 312 studies, the requests of 187 (60%) did not meet the RCR guidelines. Of these patients, 107 (57%) had no significant findings, 51 (27.3%) had incidental findings requiring further investigations, 25 (13%) had indeterminate findings of possible metastasis (of which only 3 were later proven to be metastasis) and 4 (2%) had CT proven metastasis.
Conclusion: Identifying metastatic disease is key in the management of breast cancer, however, the overuse of CT generates further investigations, and may subject patients to radiation without benefit. The RCR guidelines appear useful in identifying patients who are most likely to benefit from whole body staging CT.
Limitations: N/A
Funding for this study: N/A
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: F. Darvizeh, D. M. G. Bossi, M. Gerli, M. Alì, I. Castiglioni, D. Fazzini; Milan/IT
Purpose: To assess photon counting detector computed tomography (PCD-CT) as a single-session modality for locoregional and systemic total-body staging in breast cancer, aiming to reduce reliance on MRI, lower contrast and radiation dose, and improve patient experience and surgical planning.
Methods or Background: A PCD-CT protocol was developed including virtual non-contrast, arterial (upper abdomen), portal (thorax/abdomen), and delayed phases (abdomen/brain) with spectral reconstruction. Contrast (Iomeron 400 mg/mL) was given at 0.7 mL/kg via the arm contralateral to the tumour. Seventeen consecutive women (mean age 51; range 30–78) with biopsy-proven breast cancer scheduled for neoadjuvant chemotherapy or mastectomy were retrospectively compared with MRI. In mixed benign/malignant cases, an extra thoracic arterial phase was added. Two radiologists independently reviewed tumour, lymph nodes, and extramammary findings.
Results or Findings: PCD-CT enabled accurate detection of the primary tumour and lymphadenopathy, with diagnostic efficacy comparable to MRI. In all cases, findings matched MRI for tumour detection, size, multifocality, lymph node status, and pectoralis muscle involvement. Microcalcifications were seen only on PCD-CT, reflecting superior spatial resolution. Among 17 patients, PCD-CT revealed axillary adenopathy (n=12), internal mammary adenopathy (n=3), pectoralis fascia invasion (n=1), brain metastasis (n=1), and suspicious liver lesions (n=1). Compared with MRI+CT, PCD-CT was faster (10 vs 30 min), required a single contrast injection, and delivered ~35 mSv, ~60% lower than 64-slice CT. Supine positioning was judged more surgically favourable than prone MRI.
Conclusion: PCD-CT provides efficient, accurate locoregional and systemic staging in a single session, reducing contrast, radiation, time, and cost compared with MRI+CT.
Limitations: Small, single-centre cohort.
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethical committee was requested (study ID: 6470)