Hak Hee Kim

Bio: Hak Hee Kim is an academic researcher from University of Ulsan. The author has contributed to research in topics: Breast cancer & Mammography. The author has an hindex of 31, co-authored 189 publications receiving 3864 citations. Previous affiliations of Hak Hee Kim include Catholic University of Korea & Asan Medical Center.

Changes in cancer detection and false-positive recall in mammography using artificial intelligence: a retrospective, multireader study

Abstract: Summary Background Mammography is the current standard for breast cancer screening. This study aimed to develop an artificial intelligence (AI) algorithm for diagnosis of breast cancer in mammography, and explore whether it could benefit radiologists by improving accuracy of diagnosis. Methods In this retrospective study, an AI algorithm was developed and validated with 170 230 mammography examinations collected from five institutions in South Korea, the USA, and the UK, including 36 468 cancer positive confirmed by biopsy, 59 544 benign confirmed by biopsy (8827 mammograms) or follow-up imaging (50 717 mammograms), and 74 218 normal. For the multicentre, observer-blinded, reader study, 320 mammograms (160 cancer positive, 64 benign, 96 normal) were independently obtained from two institutions. 14 radiologists participated as readers and assessed each mammogram in terms of likelihood of malignancy (LOM), location of malignancy, and necessity to recall the patient, first without and then with assistance of the AI algorithm. The performance of AI and radiologists was evaluated in terms of LOM-based area under the receiver operating characteristic curve (AUROC) and recall-based sensitivity and specificity. Findings The AI standalone performance was AUROC 0·959 (95% CI 0·952–0·966) overall, and 0·970 (0·963–0·978) in the South Korea dataset, 0·953 (0·938–0·968) in the USA dataset, and 0·938 (0·918–0·958) in the UK dataset. In the reader study, the performance level of AI was 0·940 (0·915–0·965), significantly higher than that of the radiologists without AI assistance (0·810, 95% CI 0·770–0·850; p Interpretation The AI algorithm developed with large-scale mammography data showed better diagnostic performance in breast cancer detection compared with radiologists. The significant improvement in radiologists' performance when aided by AI supports application of AI to mammograms as a diagnostic support tool. Funding Lunit.

Pregnancy- and lactation-associated breast cancer: mammographic and sonographic findings.

Abstract: Objective. To evaluate the mammographic and sonographic findings of pregnancy-associated breast cancer. Methods. A total of 22 consecutive patients with breast cancer pathologically diagnosed during pregnancy (n = 10) or lactation (n = 12) were included in this study. The ages of the patients ranged from 26 to 49 years. Both mammography and sonography were performed on 12 patients; sonography only was performed on 7 patients; and mammography only was performed on 3 patients. Mammographic and sonographic findings were evaluated retrospectively. Results. Mammography revealed positive findings in 13 (86.7%) of 15 patients, even though all 15 patients had dense breasts.

N-butyl-2-cyanoacrylate pulmonary embolism after endoscopic injection sclerotherapy for gastric variceal bleeding.

Abstract: Purpose The purpose of this work was to describe the radiologic and clinical manifestations of n-butyl-2-cyanoacrylate pulmonary embolism (PE) after endoscopic injection sclerotherapy (EIS) for gastric variceal bleeding. Method From 1992 to 1999, the medical records of 140 patients who had undergone EIS using n-butyl-2-cyanoacrylate were reviewed for identification of respiratory symptoms and amount of injection, and their pre- and postprocedure chest radiographs were reviewed to identify PE. In patients with PE, pre- and postprocedure chest radiographs (6/6), chest CT scans (3/6), lung perfusion scans (3/6), and follow-up chest radiographs (6/6) were analyzed retrospectively. Results Radiographically evident PE was observed in 6 (4.3%) of 140 patients. In comparison with patients without emboli, these patients received a higher mean volume of injection (4.2 vs. 1.8 ml) (p = 0.0011). Four of the six patients with pulmonary emboli had respiratory symptoms. Chest radiographs and CT scans showed unusual tubular or nodular, radiopaque pulmonary emboli along the pulmonary vessels. Multiple peripheral, wedge-shaped, subsegmental perfusion defects were seen on perfusion lung scans. In five of six patients, the radiographic abnormalities showed complete or partial resolution. There were no fatalities directly associated with PE. Conclusion Radiographically evident PEs are uncommonly observed following EIS and appear to be more common in patients receiving a higher volume of liquid acrylate. Affected patients were either mildly symptomatic or asymptomatic, and there were no direct fatalities of this complication.

Diagnostic Performance of Digital versus Film Mammography for Breast-Cancer Screening

Abstract: background Film mammography has limited sensitivity for the detection of breast cancer in women with radiographically dense breasts. We assessed whether the use of digital mammography would avoid some of these limitations. methods A total of 49,528 asymptomatic women presenting for screening mammography at 33 sites in the United States and Canada underwent both digital and film mammography. All relevant information was available for 42,760 of these women (86.3 percent). Mammograms were interpreted independently by two radiologists. Breast-cancer status was ascertained on the basis of a breast biopsy done within 15 months after study entry or a follow-up mammogram obtained at least 10 months after study entry. Receiver-operating-characteristic (ROC) analysis was used to evaluate the results. results In the entire population, the diagnostic accuracy of digital and film mammography was similar (difference between methods in the area under the ROC curve, 0.03; 95 percent confidence interval, i0.02 to 0.08; P=0.18). However, the accuracy of digital mammography was significantly higher than that of film mammography among women under the age of 50 years (difference in the area under the curve, 0.15; 95 percent confidence interval, 0.05 to 0.25; P=0.002), women with heterogeneously dense or extremely dense breasts on mammography (difference, 0.11; 95 percent confidence interval, 0.04 to 0.18; P=0.003), and premenopausal or perimenopausal women (difference, 0.15; 95 percent confidence interval, 0.05 to 0.24; P=0.002). conclusions The overall diagnostic accuracy of digital and film mammography as a means of screening for breast cancer is similar, but digital mammography is more accurate in women under the age of 50 years, women with radiographically dense breasts, and premenopausal or perimenopausal women. (clinicaltrials.gov number, NCT00008346.)

Basal-Like Breast Cancer: A Critical Review

Abstract: Recent gene expression profiling of breast cancer has identified specific subtypes with clinical, biologic, and therapeutic implications. The basal-like group of tumors is characterized by an expression signature similar to that of the basal/myoepithelial cells of the breast and is reported to have transcriptomic characteristics similar to those of tumors arising in BRCA1 germline mutation carriers. They are associated with aggressive behavior and poor prognosis, and typically do not express hormone receptors or HER-2 ("triple-negative" phenotype). Therefore, patients with basal-like cancers are unlikely to benefit from currently available targeted systemic therapy. Although basal-like tumors are characterized by distinctive morphologic, genetic, immunophenotypic, and clinical features, neither an accepted consensus on routine clinical identification and definition of this aggressive subtype of breast cancer nor a way of systematically classifying this complex group of tumors has been described. Different definitions are, therefore, likely to produce variable and contradictory results that may hamper consistent identification and development of treatment strategies for these tumors. In this review, we discuss definition, heterogeneity, morphologic spectrum, relation to BRCA1, and clinical significance of this important class of breast cancer.

Epidemiology of basal-like breast cancer

Abstract: Risk factors for the newly identified “intrinsic” breast cancer subtypes (luminal A, luminal B, basal-like and human epidermal growth factor receptor 2-positive/estrogen receptor-negative) were determined in the Carolina Breast Cancer Study, a population-based, case–control study of African-American and white women. Immunohistochemical markers were used to subtype 1,424 cases of invasive and in situ breast cancer, and case subtypes were compared to 2,022 controls. Luminal A, the most common subtype, exhibited risk factors typically reported for breast cancer in previous studies, including inverse associations for increased parity and younger age at first full-term pregnancy. Basal-like cases exhibited several associations that were opposite to those observed for luminal A, including increased risk for parity and younger age at first term full-term pregnancy. Longer duration breastfeeding, increasing number of children breastfed, and increasing number of months breastfeeding per child were each associated with reduced risk of basal-like breast cancer, but not luminal A. Women with multiple live births who did not breastfeed and women who used medications to suppress lactation were at increased risk of basal-like, but not luminal A, breast cancer. Elevated waist-hip ratio was associated with increased risk of luminal A in postmenopausal women, and increased risk of basal-like breast cancer in pre- and postmenopausal women. The prevalence of basal-like breast cancer was highest among premenopausal African-American women, who also showed the highest prevalence of basal-like risk factors. Among younger African-American women, we estimate that up to 68% of basal-like breast cancer could be prevented by promoting breastfeeding and reducing abdominal adiposity.

Triple-negative breast cancer: disease entity or title of convenience?

Abstract: This Review outlines the understanding and management of triple-negative breast cancer (TNBC). TNBC shares morphological and genetic abnormalities with basal-like breast cancer (BLBC), a subgroup of breast cancer defined by gene-expression profiling. However, TNBC and BLBC tumors are heterogeneous and overlap is incomplete. Breast cancers found in BRCA1 mutation carriers are also frequently triple negative and basal like. TNBC and BLBC occur most frequently in young women, especially African Americans, and tend to exhibit aggressive, metastatic behavior. These tumors respond to conventional chemotherapy but relapse more frequently than hormone receptor-positive, luminal subtypes and have a worse prognosis. New systemic therapies are urgently needed as most patients with TNBC and/or BLBC relapse with distant metastases, and hormonal therapies and HER2-targeted agents are ineffective in this group of tumors. Poly (ADP-ribose) polymerase inhibitors, angiogenesis inhibitors, EGFR-targeted agents, and src kinase and mTOR inhibitors are among the therapeutic agents being actively investigated in clinical trials in patients with TNBC and/or BRCA1-associated tumors. Increased understanding of the genetic abnormalities involved in the pathogenesis of TNBC, BLBC and BRCA1-associated tumors is opening up new therapeutic possibilities for these hard-to-treat breast cancers.

Prognostic Value of Tumor-Infiltrating Lymphocytes in Triple-Negative Breast Cancers From Two Phase III Randomized Adjuvant Breast Cancer Trials: ECOG 2197 and ECOG 1199

Abstract: Purpose Recent studies suggest that tumor-infiltrating lymphocytes (TILs) are associated with disease-free (DFS) and overall survival (OS) in operable triple-negative breast cancer (TNBC). We seek to validate the prognostic impact of TILs in primary TNBCs in two adjuvant phase III trials conducted by the Eastern Cooperative Oncology Group (ECOG). Patients and methods Full-face hematoxylin and eosin–stained sections of 506 tumors from ECOG trials E2197 and E1199 were evaluated for density of TILs in intraepithelial (iTILs) and stromal compartments (sTILs). Patient cases of TNBC from E2197 and E1199 were randomly selected based on availability of sections. For the primary end point of DFS, association with TIL scores was determined by fitting proportional hazards models stratified on study. Secondary end points were OS and distant recurrence–free interval (DRFI). Reporting recommendations for tumor marker prognostic studies criteria were followed, and all analyses were prespecified. Results The majority of 481 evaluable cancers had TILs (sTILs, 80%; iTILs, 15%). With a median follow-up of 10.6 years, higher sTIL scores were associated with better prognosis; for every 10% increase in sTILs, a 14% reduction of risk of recurrence or death (P = .02), 18% reduction of risk of distant recurrence (P = .04), and 19% reduction of risk of death (P = .01) were observed. Multivariable analysis confirmed sTILs to be an independent prognostic marker of DFS, DRFI, and OS. Conclusion In two national randomized clinical trials using contemporary adjuvant chemotherapy, we confirm that stromal lymphocytic infiltration constitutes a robust prognostic factor in TNBCs. Studies assessing outcomes and therapeutic efficacies should consider stratification for this parameter.