RESUMEN

DNA nanotechnology plays a crucial role in precise cancer medicine. Currently, molecular logic circuits are applied to detect tumor-specific biomarkers and control the release of therapeutic drugs. However, these systems lack self-learning capabilities for intelligent diagnostics in biological samples, and their data processing capabilities are limited. Here, a molecular learning vector quantization neural network (LVQNN) model based on DNA strand displacement (DSD) technology for breast tumor diagnosis is developed. Compared to previous work, the molecular LVQNN boasts powerful computing abilities, handling high-dimensional data for intelligent cancer diagnosis. To verify the feasibility and versatility of the network, two distinct typical datasets are selected: one from a single source with cell morphology data from 569 cases, and a more extensive one spanning different populations and ages, with miRNA gene expression data from 1881 cases. By using the molecular LVQNN, diagnostic experiments are conducted on 50 and 120 public individuals from these two datasets, respectively, achieving accuracy rates of 94% and 97.5%. This study demonstrates that the LVQNN model exhibits significant advantages in breast cancer diagnosis and enhances diagnostic accuracy while introducing new approaches for intelligent cancer diagnosis, anticipated to bring significant breakthroughs and application prospects to precise cancer medicine.

RESUMEN

Regardless of whether they are benign or malignant, phyllodes tumors can behave unpredictably and aggressively. Sometimes they grow so quickly and can cause clinically different, rare, and difficult situations to manage. Because of these characteristics, they can be fatal, even if they are benign. Sometimes, emergency surgical operations may be required due to the occurrence of these conditions even before the diagnosis. We report the first case of a massive bilateral phyllodes tumor, which causes severe bleeding because of rapid growth, which resulted in emergency surgery performed after a blood transfusion. The pathological diagnosis had not yet been confirmed at the time we operated on the patient, and the postoperative pathologic examination revealed that it was a phyllodes tumor. After the successful surgical operation, the patient recovered and was discharged. In this case report, we shared the presentation and management of the emergency phyllodes tumorous phenomenon. We also conveyed our views on what could have been done differently in the management of the presented case.

RESUMEN

INTRODUCTION: Ipsilateral breast tumor recurrence (IBTR) remains a concern despite standard treatments. Advances in early detection have shifted surgical paradigms towards less invasive approaches. While repeat sentinel lymph node biopsy (rSLNB) emerges as a viable option according to the 2023 National Comprehensive Cancer Network (NCCN) guidelines, its efficacy remains uncertain. This study aimed to assess lymphatic drainage patterns in IBTR and evaluate the feasibility of rSLNB, along with analyzing oncologic outcomes. METHODS: A retrospective analysis involving 78 patients with IBTR who had prior breast-conserving surgery (BCS) with sentinel lymph node biopsy (SLNB) and adjuvant whole breast irradiation (WBI) at Samsung Medical Center was conducted. Data on patient characteristics, lymphatic mapping techniques, and oncologic outcomes were collected and analyzed. RESULTS: Among 78 patients with IBTR, 82.1 % underwent successful rSLNB, predominantly detecting lymphatic drainage to the ipsilateral axilla (80.8 %). The initial tumor location correlated significantly with failed lymphatic mapping (p = 0.019). A third event occurred in 28.8 % of invasive IBTR cases, notably associated with postmenopausal status, higher T stages, and HR(-)/HER2(-) subtype (p < 0.001). The risk of a third event increased by over 50 % within a 2-year interval post-IBTR. CONCLUSION: rSLNB in patients with IBTR, particularly for tumors initially located outside the upper-outer quadrant, demonstrated technical feasibility. The combined use of blue dye with lymphoscintigraphy may enhance rSLNB success rates. Active surveillance, especially for triple negative IBTR cases, may be important due to their aggressive nature and rapid progression potential within a short interval post-IBTR.

RESUMEN

BACKGROUND: Breast conserving surgery (BCS) is well established for the management of ductal carcinoma in situ (DCIS), but neither randomized trials nor guidelines address management of ipsilateral breast tumor recurrence (IBTR) after BCS for DCIS. PATIENTS AND METHODS: We identified women treated with BCS for DCIS who developed IBTR as a first event. Between those treated with mastectomy versus re-BCS, we compare the clinicopathologic characteristics, the use of adjuvant radiotherapy (RT) both upfront ("primary RT") and post IBTR ("secondary RT"), of tamoxifen, the rate of third events (local, regional, distant), and both breast cancer specific (BCSS) and overall survival (OS). RESULTS: Of 3001 women treated with BCS for DCIS (1978-2010), 383 developed an IBTR as a first event (1983-2023) and were treated by mastectomy (51%) versus re-BCS (49%). Compared with re-BCS, mastectomy patients at initial treatment were higher grade (74% versus 59%, p = 0.004), with more frequent primary RT (61% versus 21%, p < 0.001). Third local events were more frequent for re-BCS than mastectomy (16% versus 3%, p = 0.001), but there were no differences in breast cancer specific or overall survival. CONCLUSIONS: For isolated IBTR following BCS for DCIS and treated by mastectomy versus re-BCS (1) mastectomy was associated with less favorable initial pathology and more frequent use of primary RT, (2) re- recurrence was more frequent with re-BCS, and (3) BCSS and OS were comparable. Our data suggest a wider role for re-BCS and further study of the relationship between secondary RT and the rate of third breast events.

RESUMEN

INTRODUCTION: In academic breast surgery, ultrasound use tends to be limited to radiology departments, thus formal surgical resident training in breast ultrasound is sparse. Building on residents' ultrasound skills in our general surgery training program, we developed a novel curriculum to teach ultrasound-guided breast procedures (UGBPs), including core needle biopsy (CNB) and wire localization (WL). We hypothesized that learning UGBPs on cadavers would be preferred to learning with a breast phantom model using chicken breasts. METHODS: Residents received a 1-h lecture on breast CNB and WL followed by a 1-h hands-on laboratory session. Olives stuffed with red pimentos were used to replicate breast masses and implanted in chicken breasts and the breasts of lightly embalmed and unembalmed female cadavers. All residents practiced UGBPs with a course instructor on both models. Residents completed anonymous prelaboratory and postlaboratory surveys utilizing five-point Likert scales. RESULTS: A total of 35 trainees participated in the didactics; all completed the prelaboratory survey and 28 completed the postlaboratory survey. Participant clinical year ranged from 1 to 6. Residents' confidence in describing and performing CNBs and WLs increased significantly on postlaboratory surveys, controlling for clinical year (P < 0.001). Eighty-point seven percent preferred learning UGBPs on cadavers over phantoms most commonly citing that the cadaver was more realistic. CONCLUSIONS: Following a novel 2-h UGBP training curriculum using phantom and cadaveric models, resident confidence in describing and performing UGBPs significantly improved. Most favored the cadaveric model and reported that the course prepared them for real-life procedures.

RESUMEN

Aberrantly-active signal transducer and activator of transcription (Stat)3 has a causal role in many human cancers and represents a validated anticancer drug target, though it has posed significant challenge to drug development. A new small molecule, JKB887, was identified through virtual library screening and is predicted to interact with Lys591, Arg609 and Pro63 in the phospho-tyrosine (pTyr)-binding pocket of the Stat3 SH2 domain. JKB887 inhibited Stat3 DNA-binding activity in vitro in a time-dependent manner, with IC50 of 2.2-4.5 µM at 30-60-min incubation. It directly disrupted both the Stat3 binding to the cognate, high-affinity pTyr (pY) peptide, GpYLPQTV-NH2 in fluorescent polarization assay with IC50 of 3.5-5.5 µM at 60-90-min incubation, and to the IL-6 receptor/gp130 or Src in treated malignant cells. Treatment with JKB887 selectively blocked constitutive Stat3 phosphorylation, nuclear translocation and transcriptional activity, Stat3-regulated gene expression, and decreased viable cell numbers, cell growth, colony formation, migration, and survival in human or mouse tumor cells. By contrast, JKB887 had minimal effects on Stat1 activity, pErk1/2MAPK, pShc, pJAK2, pSrc induction, or cells that do not harbor aberrantly-active Stat3. Additionally, JKB887 inhibited growth of human breast cancer xenografts in mice. JKB887 is a Stat3-selective inhibitor with demonstrable antitumor effects against Stat3-dependent human cancers.

RESUMEN

Primary classic Hodgkin lymphoma (HL) of the breast is a rare type of breast disease. The diagnosis is mostly confirmed by an excisional biopsy. The first line of treatment commonly used for Hodgkin lymphoma is doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD). Our case report is about a 48-year-old lady who was diagnosed with bilateral breast Hodgkin lymphoma following an excisional biopsy and was treated with brentuximab vedotin plus doxorubicin, vinblastine, and dacarbazine (BV-AVD). The patient responded positively after the initiation of the regimen. There is scarce data on the classic Hodgkin lymphoma of the breast, and even with the wide use of first-line treatment using ABVD, the disease is still difficult to manage. Hence, patients with breast masses should be screened for classic HL of the breast, and larger studies are needed to establish specific treatment guidelines concerning HL of the breast to prevent relapse and other complications.

RESUMEN

Background: Endoscopic curative excision of benign breast diseases (BBDs) can preserve the cosmetic appearance of the breast. However, endoscopic surgery is not feasible, and some challenges still need to be addressed. Traditional line marker localization methods cannot visualize tumors, and the exploration of deep tumors may lead to certain risks of accidental injury. This study aimed to investigate the value of the methylene blue location (MBL) technique in endoscopic resection of deep-seated benign breast tumors. Methods: A total of 217 patients with benign deep breast tumors admitted to the Sixth Affiliated Hospital of Sun Yat-sen University between November 2017 and June 2023 met the inclusion criteria. Among them, 107 patients underwent endoscopic resection with a MBL, in which methylene blue was injected to guide the tumor resection endoscopically, whereas 110 patients underwent endoscopic resection with a skin mark location (SML), in which the tumor was located by a marking line on the skin. We compared patient characteristics, surgery-related data, complications, and cosmetic outcomes between the two groups. Results: Endoscopic breast tumor resection was successfully performed in 217 patients, none of whom had undergone open surgery. The mean operation time was significantly different between the MBL and SML groups (45.70±12.508 and 49.59±10.997 min, respectively; P=0.008<0.05), and blood loss in the MBL group was significantly reduced compared with that in the SML group (11.07±5.665 and 13.83±7.918 mL, respectively; P=0.004<0.05). There were no significant differences in drainage volume, length of hospital stay, or postoperative complications between the MBL and SML groups (P>0.05). The postoperative cosmetic outcomes of the patients were noteworthy, with no statistically significant differences between the two groups. Conclusions: The methylene blue positioning technique is safe and effective for the endoscopic treatment of deep breast tumors. It shortens operation time, reduces surgical complications, and is worthy of clinical promotion.

RESUMEN

Although research on photodynamic therapy (PDT) of malignant tumor has made considerable progress in recent years, it is a remaining challenge to extend PDT to the second near-infrared window (NIR-II) along with real-time and accurate NIR-II fluorescence imaging to determine drug enrichment status and achieve high treatment efficacy. In this work, lanthanide nanoparticles (Ln NPs)-based nanoplatform (LCR) equipped with photosensitizer Chlorin e6 (Ce6) and targeting molecular NH2-PEG1000-cRGDfK are developed, which can achieve NIR-II photodynamic therapy (PDT) and NIR-II fluorescence imaging by dual channel excitation. Under 808 nm excitation, Nd3+ in the outer layer can absorb the energy and transfer inward to emit strong NIR-II emissions (1064 and 1525 nm). Due to the low background noise of NIR-II light and the targeting effect of NH2-PEG1000-cRGDfK, LCR can recognize tiny tumor tissue (≈3 mm) and monitor drug distribution in vivo. Under 1530 nm excitation, internal Er3+ can be self-sensitized, generating intense upconversion emission (662 nm) that can effectively activate Ce6 for in vivo PDT due to the deep tissue penetration of NIR-II light. This study provides a paradigm of theranostic nanoplatform for both real-time fluorescence imaging and PDT of orthotopic breast tumor in NIR-II window.

RESUMEN

Breast conserving treatment typically involves surgical excision of tumor and adjuvant radiotherapy targeting the breast area or tumor bed. Accurately defining the tumor bed is challenging and lead to irradiation of greater volume of healthy tissues. Preoperative stereotactic body radiotherapy (SBRT) which target tumor may solves that issues. We conducted a systematic literature review to evaluates the early toxicity and cosmetic outcomes of this promising treatment approach. Secondary we reviewed pathological complete response (pCR) rates, late toxicity, patient selection criteria and radiotherapy protocols. We retrieved literature from PubMed, Scopus, Web of Science, Cochrane, ScienceDirect, and ClinicalTrials.gov. The study adhered to the PRISMA 2020 guidelines. Ten prospective clinical trials (7 phase II, 3 phase I), encompassing 188 patients (aged 18-75 years, cT1-T3 cN0-N3 cM0, primarily with ER/PgR-positive, HER2-negative status,), were analyzed. Median follow-up was 15 months (range 3-30). Treatment involved single-fraction SBRT (15-21Gy) in five studies and fractionated (19.5-31.5Gy in 3 fractions) in the rest. Time interval from SBRT to surgery was 9.5 weeks (range 1-28). Acute and late G2 toxicity occurred in 0-17% and 0-19% of patients, respectively, G3 toxicity was rarely observed. The cosmetic outcome was excellent in 85-100%, fair in 0-10% and poor in only 1 patient. pCR varied, showing higher rates (up to 42%) with longer intervals between SBRT and surgery and when combined with neoadjuvant systemic therapy (up to 90%). Preoperative SBRT significantly reduce overall treatment time, enabling to minimalize volumes. Early results indicate excellent cosmetic effects and low toxicity.

Asunto(s)

RESUMEN

Conventional benign tumor excision leaves scars on the skin surface of the breast, which is unacceptable for young patients. The feasibility and clinical results of endoscopic resection of benign breast tumors through a single axillary incision via an inflatable method were evaluated. METHODS: Clinicopathological data from 62 patients who underwent this procedure from June 2022 to July 2023 were retrospectively collected. The surgical success rate, number of tumors excised, operative time, intraoperative conditions, postoperative complications, and clinical outcomes were analyzed. RESULTS: A total of 144 benign tumors were resected, and the surgical success rate was 100% (62/62). The time required to resect tumors located in the inner quadrant with larger diameters was shorter than the time to resect tumors located in the outer quadrant with smaller tumors. Sixty-two patients experienced reduced intraoperative bleeding and did not experience skin burns, incision infections, or poor healing. However, all patients experienced varying degrees of subcutaneous emphysema during the postoperative period. Fourteen patients experienced minor local effusions, and 3 patients experienced mild upper limb dysfunction, which resolved within one month. During the six-month follow-up period, there were no cases of tumor recurrence or new tumors. The ABNSW score was above 14 points, and patient satisfaction was high. CONCLUSION: Endoscopic resection of benign breast tumors through a single incision in the axilla via the inflatable method could quickly remove lesions larger than 3.0 cm. This method also results in hidden scars and good cosmetic effects on the shape of the breast. This is a new and effective treatment for benign breast tumors. TRIAL REGISTRATION: This retrospective study was registered in the National Medical Research Registry filing system ( https://www.medicalresearch.org.cn ) (No. MR-44-22-007981) and recorded in the Medical Research Division of our hospital (No. NYXJS-22-021).

RESUMEN

Breast cancer continues to be a major global health challenge, driving the need for effective therapeutic strategies. Cisplatin, a powerful chemotherapeutic agent, is widely used in breast cancer treatment. However, its effectiveness is often limited by systemic toxicity and the development of drug resistance. This review examines the molecular factors that influence cisplatin response and resistance, offering crucial insights for the scientific community. It highlights the significance of understanding cisplatin resistance's genetic and epigenetic contributors, which could lead to more personalized treatment approaches. Additionally, the review explores innovative strategies to counteract cisplatin resistance, including combination therapies, nanoparticle-based drug delivery systems, and targeted therapies. These approaches are under intensive investigation and promise to enhance breast cancer treatment outcomes. This comprehensive discussion is a valuable resource to advance breast cancer therapeutics and address the challenge of cisplatin resistance.

RESUMEN

Lapachol (2-hydroxy-3-(3-methylbut-2-en-1-yl)naphthalene-1,4-dione) is a 1,4-naphthoquinone-derived natural product that presents numerous bioactivities and was shown to have cytotoxic effects against several human tumor cells. Indium(III) complexes with a variety of ligands also exhibit antineoplastic activity. Indium(III) complexes [In(lap)Cl2].4H2O (1), [In(lap)2Cl(Et3N)] (2), [In(lap)3]·2H2O (3) [In(lap)(bipy)Cl2] bipy = 2,2'-bipyridine (4) and [In(lap)(phen)Cl2] phen = 1,10-phenanthroline (5) were obtained with 2-hydroxy-3-(3-methylbut-2-en-1-yl)naphthalene-1,4-dione (lapachol). Crystal structure determinations for (4) and (5) revealed that the indium(III) center is coordinated to two O atoms from lapachol, two N atoms from 1,10-phenanthroline or 2,2'-bipyridine, and two chloride anions, in a distorted octahedral geometry. Although both complexes (4) and (5) interacted with CT-DNA in vitro by an intercalative mode, only 5 exhibited cytotoxicity against MCF-7 and MDA-MB breast tumor cells. 1,10-phenanthroline and complex (5) presented cytotoxic effects against MCF-7 and MDA-MB cells, with complex (5) being threefold more active than 1,10-phenanthroline on MCF-7 cells. In addition, complex (5) significantly reduced the formation of MDA-MB-231 colonies in a clonogenicity assay. The foregoing results suggest that further studies on the cytotoxic effects and cellular targets of complex (5) are of utmost relevance.

Asunto(s)

RESUMEN

Precise segmentation of breast tumors from MRI is crucial for breast cancer diagnosis, as it allows for detailed calculation of tumor characteristics such as shape, size, and edges. Current segmentation methodologies face significant challenges in accurately modeling the complex interrelationships inherent in multi-sequence MRI data. This paper presents a hybrid deep network framework with three interconnected modules, aimed at efficiently integrating and exploiting the spatial-temporal features among multiple MRI sequences for breast tumor segmentation. The first module involves an advanced multi-sequence encoder with a densely connected architecture, separating the encoding pathway into multiple streams for individual MRI sequences. To harness the intricate correlations between different sequence features, we propose a sequence-awareness and temporal-awareness method that adeptly fuses spatial-temporal features of MRI in the second multi-scale feature embedding module. Finally, the decoder module engages in the upsampling of feature maps, meticulously refining the resolution to achieve highly precise segmentation of breast tumors. In contrast to other popular methods, the proposed method learns the interrelationships inherent in multi-sequence MRI. We justify the proposed method through extensive experiments. It achieves notable improvements in segmentation performance, with Dice Similarity Coefficient (DSC), Intersection over Union (IoU), and Positive Predictive Value (PPV) scores of 80.57%, 74.08%, and 84.74% respectively.

RESUMEN

Triple-negative breast cancer (TNBC) is the most invasive type of breast cancer with high risk of brain metastasis. To better understand interactions between breast tumors with the brain extracellular matrix (ECM), a 3D cell culture model is implemented using a thiolated hyaluronic acid (HA-SH) based hydrogel. The latter is used as HA represents a major component of brain ECM. Melt-electrowritten (MEW) scaffolds of box- and triangular-shaped polycaprolactone (PCL) micro-fibers for hydrogel reinforcement are utilized. Two different molecular weight HA-SH materials (230 and 420 kDa) are used with elastic moduli of 148 ± 34 Pa (soft) and 1274 ± 440 Pa (stiff). Both hydrogels demonstrate similar porosities. The different molecular weight of HA-SH, however, significantly changes mechanical properties, e.g., stiffness, nonlinearity, and hysteresis. The breast tumor cell line MDA-MB-231 forms mainly multicellular aggregates in both HA-SH hydrogels but sustains high viability (75%). Supplementation of HA-SH hydrogels with ECM components does not affect gene expression but improves cell viability and impacts cellular distribution and morphology. The presence of other brain cell types further support numerous cell-cell interactions with tumor cells. In summary, the present 3D cell culture model represents a novel tool establishing a disease cell culture model in a systematic way.

RESUMEN

BACKGROUND: Benign breast lumps affect 10% of women in their lifetimes. Endoscopic surgery could be an alternative surgical technique for benign breast tumors because it is performed through small wounds hidden in inconspicuous areas. The aim of this study was to explore the safety and esthetic effects of endoscopic surgery in the treatment of benign breast disease. METHODS: This retrospective cohort study analyzed 363 patients with benign breast tumors from August 2021 to December 2023 in the Sixth Affiliated Hospital of Sun Yat-Sen University, of whom 118 underwent transaxillary single-port endoscopic surgery and 245 underwent traditional open surgery. Clinicopathologic characteristics, surgery type, hospital stay, and complications were analyzed to assess the effectiveness of the procedure for benign breast tumors. RESULTS: Breast tumor resection was successfully performed in 363 patients by endoscopic surgery or traditional open surgery. Endoscopic procedures demonstrated longer durations of surgery (98.54 ± 35.17 min vs. 70.28 ± 26.06 min, p < 0.01) and postoperative drainage (64.30 ± 34.92 mL vs.18.49 ± 19.86 mL, p < 0.01), but there was less blood loss. The nipple-areolar complex of the patients who underwent endoscopic resection was significantly more sensitive than the traditional open surgery group. Patients in the endoscopic group reported higher satisfaction with surgical outcome (13.10 ± 1.97 vs. 12.63 ± 1.90, p < 0.01). And there was a significant difference in the wound scar and cosmetic outcome total score between the two groups. CONCLUSION: Transaxillary single-port endoscopic surgery is effective and safe and improves postoperative nipple-areolar sensation and cosmetic outcome, as compared to the conventional technique.

RESUMEN

Background: Ipsilateral breast tumor recurrence (IBTR) following breast-conserving surgery (BCS) has been considered a risk factor for distant metastasis (DM). Limited data are available regarding the subsequent outcomes after IBTR. Therefore, this study aimed to determine the clinical course after IBTR and develop a magnetic resonance imaging (MRI)-based predictive model for subsequent DM. Methods: We retrospectively extracted quantitative features from MRI to construct a radiomics cohort, with all eligible patients undergoing preoperative MRI at time of primary tumor and IBTR between 2010 and 2018. Multivariate Cox analysis was performed to identify factors associated with DM. Three models were constructed using different sets of clinicopathological, qualitative, and quantitative MRI features and compared. Additionally, Kaplan-Meier analysis was performed to assess the prognostic value of the optimal model. Results: Among the 183 patients who experienced IBTR, 47 who underwent MRI for both primary and recurrent tumors were enrolled. Multivariate analysis demonstrated that the independent prognostic factors were human epidermal growth factor receptor 2 (HER2) status [hazard ratio (HR) =5.40] and background parenchymal enhancement (BPE) (HR =7.94) (all P values <0.01). Furthermore, four quantitative MRI features of recurrent tumors were selected through the least absolute shrinkage and selection operator (LASSO) method. The combined model exhibited superior performance [concordance index (C-index) 0.77] compared to the clinicoradiological model (C-index 0.71; P=0.006) and radiomics model (C-index 0.70; and P=0.01). Furthermore, the combined model successfully categorized patients into low- and high-risk subgroups with distinct prognoses (P<0.001). Conclusions: The clinicopathological and MRI features of IBTR were associated with secondary events following surgery. Additionally, the MRI-based combined model exhibited the highest predictive efficacy. These findings could be helpful in risk stratification and tailoring follow-up strategies in patients with IBTR.

RESUMEN

We present the case of a 52-year-old female with a giant phyllodes tumor (GPT), which was fungating through the skin that showed fleshy polypoid outgrowths. Histological analysis revealed stromal atypia, mitotic activity, and stromal overgrowth; however, the tumor border was well-defined, and malignant heterologous elements were not observed. Therefore, as some but not all malignant histological characteristics were present, we diagnosed the patient with borderline GPT. In cases of phyllodes tumor (PT) with the unique gross findings of fungation through the skin as fleshy polypoid outgrowths, caution is required for the subsequent course because even if the PT is graded as benign histologically, a malignant process can occur. Pathologists should note that the sampling of the collection site and the ambiguity of the histological grading of PT may affect the final diagnosis of GPT. It is also important to perform surgery with adequate preservation of the resected margins to control recurrence for patients with GPT.

RESUMEN

Richter transformation (RT) represents the development of intrusive lymphoma in individuals previously or concurrently diagnosed with chronic lymphocytic leukemia (CLL) and is characterized by lymph node enlargement. However, cases involving extra-nodal organ involvement as the first symptom are rare. There are no reports of RT with breast lesions as the first symptom. Nonspecific and atypical clinical manifestations represent key challenges in the accurate diagnosis and appropriate treatment of RT. This case report describes an elderly female patient who presented with breast lesions as the first RT symptom. The patient was admitted with a painless mass in the left breast. Examination revealed multiple lymphadenopathies and abnormally high white blood cell levels. The patient was diagnosed with CLL after hematological tests, assessments of bone marrow morphology, and tissue biopsy. Mammography and B-ultrasonography showed solid space-occupying lesions (BI-RADS category 5) in the left breast. Initially, the patient declined a breast biopsy and was therefore prescribed ibrupotinib treatment, which showed limited efficacy. A needle biopsy of the affected breast indicated the presence of diffuse large B-cell lymphoma. Based on auxiliary and pathological examinations and medical history, the final diagnosis was RT with breast involvement. Zanubrutinib with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone treatment provided initial control; however, the treatment strategy required adjustment because of the patient's fluctuating condition. The current status of the patient is marked as stable, showing an overall achievement of partial alleviation. The patient is in the process of receiving follow-up treatment. We also performed a comprehensive literature review on RT, with particular emphasis on its biological paradigm, prognosis implications, existing therapeutic approaches, and emerging directions in treatment modalities.

RESUMEN

BACKGROUND: Breast cancer is the most common cancer among women, and ultrasound is a usual tool for early screening. Nowadays, deep learning technique is applied as an auxiliary tool to provide the predictive results for doctors to decide whether to make further examinations or treatments. This study aimed to develop a hybrid learning approach for breast ultrasound classification by extracting more potential features from local and multi-center ultrasound data. METHODS: We proposed a hybrid learning approach to classify the breast tumors into benign and malignant. Three multi-center datasets (BUSI, BUS, OASBUD) were used to pretrain a model by federated learning, then every dataset was fine-tuned at local. The proposed model consisted of a convolutional neural network (CNN) and a graph neural network (GNN), aiming to extract features from images at a spatial level and from graphs at a geometric level. The input images are small-sized and free from pixel-level labels, and the input graphs are generated automatically in an unsupervised manner, which saves the costs of labor and memory space. RESULTS: The classification AUCROC of our proposed method is 0.911, 0.871 and 0.767 for BUSI, BUS and OASBUD. The balanced accuracy is 87.6%, 85.2% and 61.4% respectively. The results show that our method outperforms conventional methods. CONCLUSIONS: Our hybrid approach can learn the inter-feature among multi-center data and the intra-feature of local data. It shows potential in aiding doctors for breast tumor classification in ultrasound at an early stage.

Asunto(s)