Identification and Validation of DEPDC1B as an Independent Early Diagnostic and Prognostic Biomarker in Liver Hepatocellular Carcinoma

Fan, Xiaoyan; Wen, Jiaxin; Bao, Lei; Gao, Fei; You, Li; He, Dongwei

doi:10.3389/fgene.2021.681809

Cited by 5 publications

(5 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cancers with similar prognosis are generally clustered together, while those with differing prognostic profiles have more distinct clustering patterns with respect to one another. For example, LIHC has a relatively poor prognosis for most patients [ 63 ], while BRCA and KIRC have higher survival rates [ 64 , 65 ]. Accordingly, BRCA and KIRC are generally clustered together, while LIHC is more distinct from the two.…”

Section: Resultsmentioning

confidence: 99%

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication

Azher¹,

Suvarna²,

Chen³

et al. 2023

BioData Mining

View full text Add to dashboard Cite

Background Deep learning models can infer cancer patient prognosis from molecular and anatomic pathology information. Recent studies that leveraged information from complementary multimodal data improved prognostication, further illustrating the potential utility of such methods. However, current approaches: 1) do not comprehensively leverage biological and histomorphological relationships and 2) make use of emerging strategies to “pretrain” models (i.e., train models on a slightly orthogonal dataset/modeling objective) which may aid prognostication by reducing the amount of information required for achieving optimal performance. In addition, model interpretation is crucial for facilitating the clinical adoption of deep learning methods by fostering practitioner understanding and trust in the technology. Methods Here, we develop an interpretable multimodal modeling framework that combines DNA methylation, gene expression, and histopathology (i.e., tissue slides) data, and we compare performance of crossmodal pretraining, contrastive learning, and transfer learning versus the standard procedure. Results Our models outperform the existing state-of-the-art method (average 11.54% C-index increase), and baseline clinically driven models (average 11.7% C-index increase). Model interpretations elucidate consideration of biologically meaningful factors in making prognosis predictions. Discussion Our results demonstrate that the selection of pretraining strategies is crucial for obtaining highly accurate prognostication models, even more so than devising an innovative model architecture, and further emphasize the all-important role of the tumor microenvironment on disease progression.

show abstract

Section: Resultsmentioning

confidence: 99%

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication

Azher¹,

Suvarna²,

Chen³

et al. 2023

BioData Mining

View full text Add to dashboard Cite

show abstract

“…Cancers with similar prognosis are generally clustered together, while those with differing prognostic profiles have more distinct clustering patterns with respect to one another. For example, LIHC has a relatively poor prognosis for most patients [61], while BRCA and KIRC have higher survival rates [62, 63]. Accordingly, BRCA and KIRC are generally clustered together, while LIHC is more distinct from the two.…”

Section: Resultsmentioning

confidence: 99%

Assessment of Emerging Pretraining Strategies in Interpretable Multimodal Deep Learning for Cancer Prognostication

Azher¹,

Suvarna²,

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning models have demonstrated the remarkable ability to infer cancer patient prognosis from molecular and anatomic pathology information. Studies in recent years have demonstrated that leveraging information from complementary multimodal data can improve prognostication, further illustrating the potential utility of such methods. Model interpretation is crucial for facilitating the clinical adoption of deep learning methods by fostering practitioner understanding and trust in the technology. However, while prior works have presented novel multimodal neural network architectures as means to improve prognostication performance, these approaches: 1) do not comprehensively leverage biological and histomorphological relationships and 2) make use of emerging strategies to "pretrain" models (i.e., train models on a slightly orthogonal dataset/modeling objective) which may aid prognostication by reducing the amount of information required for achieving optimal performance. Here, we develop an interpretable multimodal modeling framework that combines DNA methylation, gene expression, and histopathology (i.e., tissue slides) data, and we compare the performances of crossmodal pretraining, contrastive learning, and transfer learning versus the standard procedure in this context. Our models outperform the existing state-of-the-art method (average 11.54% C-index increase), and baseline clinically driven models. Our results demonstrate that the selection of pretraining strategies is crucial for obtaining highly accurate prognostication models, even more so than devising an innovative model architecture, and further emphasize the all-important role of the tumor microenvironment on disease progression.

show abstract

“…Recently, Lai et al, reported that DEPDC1B was identified as a key regulator of bladder cancer development and could be used as a potential therapeutic target for bladder cancer therapy [ 17 ]. Moreover, Fan et al indicated that DEPDC1B as an independent early diagnostic and prognostic biomarker for hepatocellular carcinoma [ 25 ]. Therefore, DEPDC1B may be a key regulator of cancer progression, and the detection of its expression level had important clinical value for diagnosis and prognosis of patients.…”

Section: Discussionmentioning

confidence: 99%

DEPDC1B collaborates with GABRD to regulate ESCC progression

et al. 2022

View full text Add to dashboard Cite

Background Esophageal squamous cell carcinoma (ESCC) is the leading cause of cancer-related death worldwide with a poor prognosis. Given that DEPDC1B plays a key role in multiple cancers, the role of this molecule in ESCC was explored to identify potential targets for ESCC patients. Method The expression level of DEPDC1B in ESCC was revealed based on the TCGA database and immunohistochemical experiments on clinical tissues. The correlation between DEPDC1B and survival of ESCC patients was analyzed by Kaplan–Meier method. Small hairpin RNA (shRNA)-mediated silencing of DEPDC1B expression in ESCC cells and performed a series of in vitro and in vivo functional validations. Result DEPDC1B was overexpressed in ESCC. High expression of DEPDC1B was significantly negatively correlated with overall survival in patients with ESCC. Moreover, knockdown of DEPDC1B inhibited ESCC cell proliferation, clone formation, migration, tumor formation and promoted apoptosis. Furthermore, knockdown of DEPDC1B leaded to significant downregulation of GABRD in ESCC cells. Meanwhile, GABRD expression was upregulated in ESCC, and its silencing can inhibit the proliferation and migration of the tumor cells. Interestingly, there was a protein interaction between DEPDC1B and GABRD. Functionally, GABRD knockdown partially reversed the contribution of DEPDC1B to ESCC progression. In addition, GABRD regulated ESCC progression may depend on PI3K/AKT/mTOR signaling pathway. Conclusion DEPDC1B collaborated with GABRD to regulate ESCC progression, and inhibition of this signaling axis may be a potential therapeutic target for ESCC.

show abstract

Identification and Validation of DEPDC1B as an Independent Early Diagnostic and Prognostic Biomarker in Liver Hepatocellular Carcinoma

Cited by 5 publications

References 37 publications

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication

Assessment of Emerging Pretraining Strategies in Interpretable Multimodal Deep Learning for Cancer Prognostication

DEPDC1B collaborates with GABRD to regulate ESCC progression

Contact Info

Product

Resources

About