Dataset Information

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR⁺/HER2^- breast cancer.

ABSTRACT:

Background

Breast cancer has the highest incidence and mortality rates among women worldwide. Hormone receptor (HR)⁺/human epidermal growth factor receptor 2 (HER2)^- breast cancer is the most common molecular subtype, accounting for 50-79% of breast cancers. Deep learning has been widely used in cancer image analysis, especially for predicting targets related to precise treatment and patient prognosis. However, studies focusing on therapeutic target and prognosis predicting in HR⁺/HER2^- breast cancer are lacking.

Methods

This study retrospectively collected hematoxylin and eosin (H&E)-stained slides of HR⁺/HER2^- breast cancer patients between January 2013 and December 2014 at Fudan University Shanghai Cancer Center (FUSCC) and scanned to generate whole-slide images (WSIs). Then, we built a deep-learning-based workflow to train and validate model to predict clinicopathological features, multi-omics molecular features and prognosis; the area under the curve (AUC) of the receiver operating characteristic (ROC) and the concordance index (C-index) of the test set were used to assess model effectiveness.

Results

A total of 421 HR⁺/HER2^- breast cancer patients were included in our study. Regarding clinicopathological features, grade III could be predicted with an AUC of 0.90 [95% confidence interval (CI): 0.84-0.97]. Regarding somatic mutations, TP53 and GATA3 mutation could be predicted with AUCs of 0.68 (95% CI: 0.56-0.81) and 0.68 (95% CI: 0.47-0.89), respectively. Regarding gene set enrichment analysis (GSEA) pathways, the G2-M checkpoint pathway was predicted with an AUC of 0.79 (95% CI: 0.69-0.90). Regarding markers of immunotherapy response, intratumoral tumor-infiltrating lymphocytes (iTILs), stromal tumor-infiltrating lymphocytes (sTILs), CD8A, and PDCD1 were predicted with AUCs of 0.78 (95% CI: 0.55-1.00), 0.76 (95% CI: 0.65-0.87), 0.71 (95% CI: 0.60-0.82), and 0.74 (95% CI: 0.63-0.85), respectively. In addition, we found that the integration of clinical prognostic variables and deep features of images can improve the stratification of patient prognosis.

Conclusions

Using a deep-learning-based workflow, we developed models to predict the clinicopathological features, multi-omics features and prognosis of patients with HR⁺/HER2^- breast cancer using pathological WSIs. This work may contribute to efficient patient stratification to promote the personalized management of HR⁺/HER2^- breast cancer.

SUBMITTER: Hu J

PROVIDER: S-EPMC10267923 | biostudies-literature | 2023 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR<sup>+</sup>/HER2<sup>-</sup> breast cancer.

Hu Jia J Lv Hong H Zhao Shen S Lin Cai-Jin CJ Su Guan-Hua GH Shao Zhi-Ming ZM

Journal of thoracic disease 20230523 5

<h4>Background</h4>Breast cancer has the highest incidence and mortality rates among women worldwide. Hormone receptor (HR)<sup>+</sup>/human epidermal growth factor receptor 2 (HER2)<sup>-</sup> breast cancer is the most common molecular subtype, accounting for 50-79% of breast cancers. Deep learning has been widely used in cancer image analysis, especially for predicting targets related to precise treatment and patient prognosis. However, studies focusing on therapeutic target and prognosis pr ...[more]

PMID: 37324098

Similar Datasets

Project description:BackgroundCyclin-dependent kinase 4/6 inhibitor (CDK4/6) therapy plus endocrine therapy (ET) is an effective treatment for patients with hormone receptor-positive/human epidermal receptor 2-negative metastatic breast cancer (HR+/HER2- MBC); however, resistance is common and poorly understood. A comprehensive genomic and transcriptomic analysis of pretreatment and post-treatment tumors from patients receiving palbociclib plus ET was performed to delineate molecular mechanisms of drug resistance.MethodsTissue was collected from 89 patients with HR+/HER2- MBC, including those with recurrent and/or metastatic disease, receiving palbociclib plus an aromatase inhibitor or fulvestrant at Samsung Medical Center and Seoul National University Hospital from 2017 to 2020. Tumor biopsy and blood samples obtained at pretreatment, on-treatment (6 weeks and/or 12 weeks), and post-progression underwent RNA sequencing and whole-exome sequencing. Cox regression analysis was performed to identify the clinical and genomic variables associated with progression-free survival.ResultsNovel markers associated with poor prognosis, including genomic scar features caused by homologous repair deficiency (HRD), estrogen response signatures, and four prognostic clusters with distinct molecular features were identified. Tumors with TP53 mutations co-occurring with a unique HRD-high cluster responded poorly to palbociclib plus ET. Comparisons of paired pre- and post-treatment samples revealed that tumors became enriched in APOBEC mutation signatures, and many switched to aggressive molecular subtypes with estrogen-independent characteristics. We identified frequent genomic alterations upon disease progression in RB1, ESR1, PTEN, and KMT2C.ConclusionsWe identified novel molecular features associated with poor prognosis and molecular mechanisms that could be targeted to overcome resistance to CKD4/6 plus ET.Trial registrationClinicalTrials.gov, NCT03401359. The trial was posted on 18 January 2018 and registered prospectively.

Project description:BackgroundThis study aims to analyze potential differences in clinicopathology, efficacy of neoadjuvant therapy (NAT), and clinical outcome among HER2-null, HER2-ultralow and HER2-low breast cancers.MethodsConsecutive cases of HER2-negative breast cancer that received NAT were included. They were classified as HER2-null (no staining), HER2-ultralow (incomplete faint staining in ≤ 10% of tumour cells) and HER2-low (HER2-1 + or HER2-2+, in situ hybridisation negative). Subgroup analysis was performed based on the HER2 expression level.ResultsOut of 302 patients, 215 (71.19%) were HER2-low, 59 (19.54%) were HER2-ultralow, and 28 (9.27%) were HER2-null. In comparison to the HER2-ultralow group, the HER2-low group exhibited higher expression frequencies of ER (p < 0.001), PR (p < 0.001), and AR (p = 0.004), along with a greater prevalence of the luminal subtype (p < 0.001). The HER2-ultralow group also demonstrated a higher prevalence of lymph node metastasis compared to the HER2-null group (p = 0.026). Varied rates of pathologic complete response (pCR) were observed among the three subgroups: HER2-null, HER2-ultralow, and HER2-low, with rates of 35.71%, 22.03%, and 12.56%, respectively. Only the HER2-low subgroup exhibited a significant difference compared to HER2-null (p = 0.001). Despite variations in pCR rates, the three subgroups exhibited comparable disease-free survival (DFS) (p = 0.571). Importantly, we found HER2-low patients with better treatment response (RCB-0/I) exhibited significantly better DFS than those with significant residual disease (RCB-II/III) (P = 0.036). The overall rate of HER2 immunohistochemical score discordance was 45.24%, mostly driven by the conversion between HER2-0 and HER2-low phenotype. Notably, 32.19% of cases initially classified as HER2-0 phenotype on baseline biopsy were later reclassified as HER2-low after neoadjuvant therapy, and it is noteworthy that 22 out of these cases (78.57%) originally had an HER2-ultralow status in the pretreatment biopsy sample.ConclusionsOur results demonstrate the distinct clinicopathological features of HER2-low and HER2-ultralow breast tumors and confirm that RCB is an effective predictor of prognosis in HER2-low populations for the first time. Notably, our findings demonstrate high instability in both HER2-low and HER2-ultralow expression from the primary baseline biopsy to residual disease after NAT. Furthermore, this study is the first to investigate the clinicopathological feature and the effectiveness of NAT for HER2-ultralow breast cancer.

Project description:BackgroundThyroid dysfunction is common in patients with kidney disease. However, the relationship between thyroid dysfunction and idiopathic membranous nephropathy (IMN) remains unclear. This retrospective study aimed to investigate the clinicopathological characteristics and prognosis of patients with IMN and thyroid dysfunction compared to patients with IMN and without thyroid dysfunction.MethodsA total of 1052 patients with IMN diagnosed by renal biopsy were enrolled in this study, including 736 (70%) with normal thyroid function and 316 (30%) with abnormal thyroid function. We analyzed the clinicopathological features and prognostic data between the two groups, using propensity score matching (PSM) to reduce the bias. Logistic regression analysis was performed to investigate the risk factors for IMN combined with thyroid dysfunction. Kaplan-Meier curves and Cox regression analysis were used to evaluate the association between thyroid dysfunction and IMN.ResultsPatients with IMN and thyroid dysfunction exhibited more severe clinical features. Female sex, lower albumin level, higher D-dimer level, severe proteinuria, and decreased estimated glomerular filtration rate were predictors of thyroid dysfunction in patients with IMN. After PSM, 282 pairs were successfully matched. Results from the Kaplan-Meier curves indicated that the thyroid dysfunction group had a lower complete remission rate (P = 0.044), higher relapse rate (P < 0.001), and lower renal survival rate (P = 0.004). The multivariate Cox regression analysis revealed that thyroid dysfunction was an independent risk factor for complete remission [hazard ratio (HR) = 0.810, P = 0.045], relapse (HR = 1.721, P = 0.001), and composite endpoint event (HR = 2.113, P = 0.014) in IMN.ConclusionsThyroid dysfunction is relatively common in patients with IMN, and the clinical indicators are more severe in these patients. Thyroid dysfunction is an independent risk factor for poor prognosis in patients with IMN. More attention should be paid to thyroid function in patients with IMN.

Dataset Information

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR⁺/HER2^- breast cancer.

Background

Methods

Results

Conclusions

Publications

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR<sup>+</sup>/HER2<sup>-</sup> breast cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR+/HER2- breast cancer.

Background

Methods

Results

Conclusions

Publications

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR<sup>+</sup>/HER2<sup>-</sup> breast cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Prediction of clinicopathological features, multi-omics events and prognosis based on digital pathology and deep learning in HR⁺/HER2^- breast cancer.