Abstract
Introduction:
Breast cancer is the highest cause of female cancer deaths worldwide. Africa bears the brunt of this devastating disease mainly due to a lack of awareness and late presentation. Recently, a new cohort of patients in some jurisdictions in Africa have presented with small nonpalpable breast tumors due to early detection or following neoadjuvant chemotherapy.
Aim:
This study documented the initial experience of wire-guided localization of nonpalpable breast tumors in a Ghanaian tertiary hospital, used to facilitate the achievement of negative surgical margins.
Methods:
This was a retrospective evaluation and analysis of clinical, radiological, and histopathological data of 45 patients who had image-guided wire localization of nonpalpable lumps immediately prior to surgical excision at the Korle Bu Teaching Hospital over more than a 4-year period. The study evaluated the preprocedural radiological diagnosis, tumor size, histology, and completeness of resection.
Results:
Median age at presentation was 50 years. Clinical indications of these nonpalpable lesions included 13.3% post neoadjuvant chemotherapy and 40.0% of chemotherapy naive histologically diagnosed breast cancers. The median size of the excised lesions was 13 mm. Excision was associated with clear margins in most cases. Up to 53.3% of the lesions were malignant, out of which invasive ductal carcinoma NST was the commonest histology.
Conclusion:
Ultrasound-guided wire localization has proven to be a beneficial tool in breast-conserving surgery in an LMIC environment. More than half the pathologies localized were malignant, with 96% showing clear margins.
Introduction
Breast cancer remains the commonest malignancy in women worldwide. According to GLOBOCAN data reported an incidence of approximately 2.3 million new breast cancer cases accounting for 23.8% of cancer cases in 2022 and a mortality of 667 000, representing 6.8% of all cancer deaths globally. Ghana for the same year recorded an incidence of 5026 new breast cancer cases (18.4% of all newly diagnosed cancers), and a mortality of 2369—giving a higher than average mortality rate of 13.2% of all cancer mortalities.1,2 At the Korle Bu Teaching Hospital (KBTH), Ghana’s largest tertiary hospital, breast cancer is the commonest cancer diagnosed, representing 29% of all cancers. 3 Unpublished data from this hospital reports approximately 600 new breast cancer cases diagnosed every year. In Ghana breast cancer is typically diagnosed late. Approximately 60% of diagnosed cases are advanced at presentation, requiring mastectomy when surgery is indicated.4,5 However, there is now data suggesting some increasing early-stage diagnosis. 6
In Ghana, fear of mastectomy is a major contributing factor for late presentation and non-compliance with treatment. 7 Despite the lack of a comprehensive national breast cancer screening policy, recent years have seen an increase in breast cancer awareness activities and opportunistic screening in Ghana with an attendant increase in detection of early disease. Therefore, the possibility of breast-conserving surgery (BCS) in the appropriate group of cases will contribute to the easy assimilation of the treatment plan. 8 In addition to the psychological, body image and other quality of life benefits derived from BCS.9 -11 Studies have not shown significant difference in the prognosis and survival of patients who had mastectomy compared with BCS. 12 This is also relevant for reasonably large tumors that respond to neoadjuvant chemotherapy and BCS achieved. Diagnosing these breast cancers early, has brought on the attendant challenge of being able to accurately excise nonpalpable lesions, with wire localization techniques helping to bridge that gap. Locally our experience with localization devices has transcended our initial use of fabricated steel wires to the current use of generic commercially produced wires. This has been in conjunction with an improved ability of our interventional radiologists to perform image-guided biopsies on sub centimetric lesions, while still keeping our overhead costs low at 80 USD per wire-guided localization (WGL) procedure and BCS at approximately 300 USD.
Interventional Radiology is a limited and slowly evolving sub-specialty in sub-Saharan Africa and as the only facility providing this service in the country, it became imperative that we appraise our work and determine if the procedure has been beneficial and successful. The authors hope that highlighting the findings of this study will encourage the utilization of this invaluable tool in the management of nonpalpable breast cancers at other breast cancer treatment centers in the country and subregion. This study evaluated the preprocedural clinical and radiological findings of the lumps and determined the histological characteristics and successfulness of the excision facilitated by WGL.
Methods
Study design and site
The reporting of this study conforms to the STROBE statement for observational studies. 13 This was a retrospective cross-sectional study conducted at the Korle Bu Teaching Hospital (KBTH) from January 2020 to December 2024. All clients who had had wire-guided localization and subsequent surgical excision of their breast lesions from January 2020 to December 2024 were included. The inclusion criteria included patients older than 18 years with nonpalpable breast lumps in whom excision was indicated whether histologically proven malignant or not malignant. It also included patients who may or may not have had chemotherapy. The exclusion criteria in cooperated patients whose histopathology reports were incomplete or not available. Records of all patients who had wire localization was extracted from the procedure book at the department of radiology. A total of 50 cases were collated over the study period. Out of this 45 had complete data which could be analyzed.
This preliminary information was used to access their clinical data in the hospital database to extract patient demographic, clinical and histopathological data. Details collated included; age of patient, clinical stage, indication for procedure, Breast Imaging-Reporting and Data System (BIRADS) category of lesion, affected side, specimen size, histological diagnosis, and state of resection margins—including complete resection with negative or positive margins. Complications related to wire localization including hematoma formation, wire migration or re-excision rates were recorded.
Study site
The Korle Bu Teaching Hospital is Ghana’s largest and oldest tertiary hospital. It is the largest breast cancer referral site in the country, seeing up to 600 new breast cancer cases yearly; these include referrals from the Southern sector of the Ghana as well as from the West African Subregion. The hospital also boasts of a dedicated breast clinic which manages both referred and non-referred cases. We have a dedicated breast surgical unit, a well-equipped radiology, pathology, radiotherapy, oncology, and nuclear medicine center which provide comprehensive cancer care.
Margin positivity
Negative margins are defined as no ink on tumor. Positive margins defined as ink on invasive cancer or ductal carcinoma in situ (DCIS).
Close margins defined histologically as margins measuring 2 mm or less.
Data analysis
The data was cleaned and fed into the IBM Statistical Package for the Social Sciences (SPSS) version 26 for analysis. Descriptive statistics were used to summarize the data—reported as median and percentages.
Localization technique
All patients had had prior ultrasound-guided core needle biopsies by the interventional radiologists. Due to non-availability no clip placement was performed for patients who had neoadjuvant chemotherapy.
The wire localization procedure was performed by 2 consultant radiologists with 5 and 15 years’ experience. In preparation for the procedure the location of the lesion was verified a day prior on ultrasound. An informed consent was obtained, and the procedure explained to the patient on the morning of the procedure. At the start of the procedure the patient was positioned supine and draped. Under aseptic conditions and ultrasound guidance 5 to 10 mL of 2% injection lidocaine was used to infiltrate the skin and breast tissue around the mass after which a Q type disposable localization wire device was placed into the lesion, Figure 1. Localization was always intralesional and performed in the radiology department without the surgeon present.

Gray scale ultrasound images acquired during ultrasound-guided wire localization procedure. (A) and (B) show a 1.4 × 0.8 cm residual mass following neoadjuvant chemotherapy. (C) Partly visualized localization wire traversing the mass (yellow arrow).
A loose sterile dressing was placed over the wire and taped down. The size and location of the lesion was documented with a graphical representation of distance along the wire and perpendicular distance from the lesion to skin (Figure 2). Each patient was subsequently taken to the operating room for same day excision of the lesion. Under general anesthesia, a wide local excision was done using the accompanying radiology notes as a guide. Prophylactic administration of antibiotic using IV Amoxicillin/Clavulanic acid 1.2 g was given as a stat dose at induction of anesthesia and repeated at 8 and 16 hours after surgery. When indicated a standard axillary lymph node dissection was done, for histologically confirmed cancers. The wire was left in situ post resection and appropriate marking of the specimen done prior to transferring the specimen to pathology.

Sample of post procedure documentation.
Results
A total of 45 specimens of 45 women who had wire localization was analyzed. The median age of the participants was 50 (interquartile range, 26-83) years. The median age of the participants with malignant lesions was 51 years. Clinically all the lesions were nonpalpable at the time of the wire localization procedure. Twenty-four of the 45 cases were malignant with 13.3% (6/45) having had neoadjuvant chemotherapy as they were initially larger clinically palpable tumors (T3 and T4) that had complete clinical response becoming no longer palpable and 40.0% (18/45) were chemotherapy naive histologically diagnosed breast cancers. The rest of the cases 46.7% (21/45) were radiologically indeterminate lesions and radiologically nonmalignant lesions found in the contralateral breast of in breast cancer patients, shown in Table 1.
Patient demographics and clinical characteristics.
The preprocedural BIRADS categories based on mammogram and ultrasound, included BIRADS 3 in 26.7% (12/45), BIRADS 4 in 40.0 % (18/45), BIRADS 5 in 20.0 % (9/45) of cases and BIRADS 6 in 13.3% (6/45) of cases.
Preprocedural clinical staging of the malignant tumors using the TMN classification was done on only the malignant lesions, 24/45; the staging for the neoadjuvant chemotherapy group was based on their prechemotherapy state, as follows; 12.5 % (3/24) as T1, 58.3% (14/24) of T2 cases, 16.7% (4/24) of T3 cases and 4.2 % (1/24) of T4 cases. Nodal staging ranged from 37.5% (9/24) N0 cases, 50.0% (12/24) N1 cases and 4.2% (1/24) of N2 cases. There were no cases of distant metastases. These correlated with clinical staging; 8.3 % (2/24) were stage 1, 37.5% (9/24) stage IIA, 16.7% (4/24) stage IIB, 25% (6/24) stage IIIA and 4.1% (1/24) stage IIIB. There were 2 cases of unknown clinical stage.
A total of 25% (6/24) of the malignant lesions had neoadjuvant chemotherapy.
The histological yield included 53.3% (24/45) malignancies out of which 4.1% (1/24) demonstrated no significant residual tumor following neoadjuvant chemotherapy. The histological types are listed in Table 2. The remaining 44.4% (20/45) were benign lesion, these included fibroadenoma and fibrocystic lesions.
Characteristics of excised malignant lesions.
Majority of the lesions were right sided in 57.8% (26/45) and left sided in 42.2 % (19/55). Out of this 58.3% of the malignant lesions were right sided and 41.7% left sided.
The excised lesions were determined histologically post-excision at pathology to have a median size of 13 mm (range, 1-33 mm) out of the 23/45 cases documented. In 3/45 cases there was pathological complete response. Post resection samples showed varying degrees of pathological clear margins, these included, less than 2 mm in 29.6% (8/27), 2 to 5 mm clear margins in 25.9% (7/27), 6 to 10 mm clear margins in 11.1% (3/27), 11 to 20 mm in 22.2% (6/27). In 7.4% (2/27), there was more than 21 mm clear margins. However, in 3.7% (1/27) the specimen showed involved margins.
Findings related to the hormone receptor status are summarized in Table 2.
There were 3 cases of minor hematoma, however, no wire migration or dislodgement occurred in association with the procedure in this study.
Discussion
This study evaluated the operationalization of WGL as a tool to facilitate breast-conserving surgery for nonpalpable breast cancers in a low-middle income country (LMIC) and found it to be beneficial, achieving 96% clear margins. This became essential as we continue to embrace the paradigm shift of early-stage breast cancer diagnoses in sub-Saharan Africa.
This study reports a young median age of 51 years for the malignant lesions relative to the Hendrick et al study which demonstrated a higher median age of 61 for their white population.16,17 This agrees with known racial epidemiological differences among breast cancer patients. 17
Just over half of this study’s participants had a histologically confirmed malignant diagnosis prior to WGL with this number increasing to more than 70% when the pre-histology BIRADS category of 4, 5, and 6 was used. Localization criteria over the years have been expanded to include nonmalignant aetiologies in the contralateral breast or as the only lesion due to reasons such as patient anxiety and risk of inadequate sampling for small lesion. 18 Ghana like many sub-Saharan African countries do not have robust screening and surveillance systems which may limit the ability to solely depend on surveillance, as such there is a tendency to excise BIRADS 3 and even some obviously benign lesions. The Khare et al 18 study gave credence to localization of nonmalignant lesions in their study which included some benign nonpalpable lesions, their findings however differed from ours with a smaller percentage of benign lesion.
The majority (75%) of the localized malignant lesions at presentation met the inclusion criteria, while the rest become only nonpalpable following neoadjuvant chemotherapy. This emphasizes the growing number of early-stage breast cancer in our jurisdiction and corroborating further the need for this procedure. Clinical staging influences the treatment modality. Most of the malignant lesions were clinically stage I with under 30% of stage IIIA and IIIB requiring neoadjuvant chemotherapy.
Of importance was the size of the lesion. The lesion size in our study ranged between 1 and 33 mm, as determined at pathology. Similar sizes were reported by Dimitrovska et al 19 also using Hook wire localization on ultrasound and mammography localized lesions from less than 5 to 20 mm.
Evidence of a successful surgical excision with WGL includes the degree of completeness of resection without a requirement for re-excision and acceptability of the oncoplastic outcome for the surgeon and patient, the latter however, was beyond the scope of this study. Approximately 30% of cases had close margins, nevertheless, the vast majority of these cases had margins 2 mm or more, with 3.7% (1/24) showing positive margins. Definitions and implications of margin positivity continue to evolve. ASCO/America Society of Breast Surgeons guidelines recommend no tumor on ink for invasive cancer as adequate. 15 The Moran et al 15 Consensus Guideline for margins in breast-conserving surgery metanalysis among their conclusions did not find a clear relationship between the degree of clear margin width and ipsilateral breast tumor recurrence (IBTR) rate, consequently, extent or size of a negative margin was not an indication for repeat surgery or increased radiation dose to the tumor bed. On the contrary, there is abundant evidence of increased IBTR rate in the presence of margin positively.15,20 The case of positive margin in this study had re-excision.
Over the years several studies have attempted to compare the adequacy of wire versus non-wire localization techniques like radioactive occult lesion localization (ROLL), radioactive seed localization (RSL) and non-radioactive techniques like radiofrequency-guided localization, Cryo-assisted localization and intraoperative ultrasound-guided localization (USGL) among others. A network metanalysis by Davey et al, 21 concluded that WGL is still the most prevalent modality used worldwide probably because of the relatively low attendant set up cost and ease of use, therefore ideal for LMICs. Their study showed that cryo-assisted localization had the highest margin positivity rate followed by WGL, with the lowest seen with USGL.18,21 USGL was also associated with the lowest rate of reoperation when compared with WGL, an advantage not explicitly derived from the other modalities. In comparison, only a small percentage of our cases had close or positive margins which may have been a product of our low study numbers. The study by Davey et al also brought to bear some of the various advantages over WGL, for example, WGL limited the site of initial incision unlike some of the others, with the requirement of strict scheduling on the day of surgery while some others can be decoupled from the day of surgery and hence patients from areas without the procedural expertise could still benefit from the procedure. 21 USGL much like WGL is also potentially low cost but requires intraoperative localization real-time with ultrasound which again is heavily dependent on radiology-surgical scheduling. USGL can be adopted and used in our setting given the low cost. Drozgyik et al in their randomized controlled study also tested the superiority of ROLL over WGL techniques. Their study demonstrated a clear advantage of ROLL in providing axillary sentinel lymph node guidance. 22 Just like the Davey et al study, Drozgyik et al also highlighted several positive patient attributes, such as reduced pain and discomfort, and relatively faster procedural time, the latter also further depicted in a much larger Egyptian study by Elzohery et al. 23 A recent study by Parisi et al 24 adds to the discussion of alternative techniques. In their critical work comparing a combined novel LOCALizer and ultrasound localization with WGL, the former showed superior oncological radicality and overall patient satisfaction. The Parisi et al 25 group in an earlier study again showed the effectiveness of combining the LOCALizer, with Intraoperative ultrasound (IOUS).
These other localization techniques despite their flexibility come with huge setup costs which would be prohibitive in resource limited settings. The use of radiation in some cases also comes with the requirement of precise safety and quality assurance practices which are not always tenable in resource poor settings like ours. 21
Overall, our experience with WGL over the past 4 years has been positive despite these initial small numbers. All of the lesions with the exception of one demonstrated sufficiently clear margins at histology, requiring re-excision.
The authors of this study in highlighting our experience and success with WGL prior to surgical excision recommend this low cost and easily adoptable procedure to colleagues across the country and the subregion, as an adjunct to breast-conserving surgery to ensure that their patients benefit from breast-conserving techniques. It would be worth researching into and adopting other more novel localization techniques in the future.
Limitations
This study was limited by the small sample size and the retrospective nature limited our ability to acquire complete data for all variables. We propose a larger prospective study as the detection of early disease increases. Our limited experience with the newer novel techniques also restricted the scope of our study.
Conclusion
Almost half of the lesions localized were malignant, majority being the invasive ductal carcinomas NST type. Ninety-six percent of lesions excised had sufficiently clear margins. Save for minor hematomas no significant complications were recorded. WGL is recommended as an effective tool for localizing nonpalpable breast lesions making breast conservation a reality in LMICs.
Footnotes
Acknowledgements
The authors thank Collins Asante and Sylvia Tetteh for their contribution with data collection.
ORCID iDs
Ethical considerations
The study was approved by the institutional review board of the Korle Bu Teaching with protocol identification number KBTH-STC-00076/2023.
Consent for publication
All authors approved the manuscript and gave consent for publication.
Author contributions
Hafisatu Gbadamosi: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.
Josephine Nsaful: Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.
Yaw Boateng Mensah: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – review & editing.
Florence Dedey: Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Visualization, Writing – review & editing.
Joe-Nat Clegg-Lamptey: Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Validation, Writing – review & editing.
Simpson Mensah: Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Validation, Visualization, Writing – review & editing.
Dinah Essah: Conceptualization, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – review & editing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
All data will be made available upon reasonable request to the corresponding author.
