{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Wang X"],"funding":["BLRD VA","NCATS NIH HHS","NIBIB NIH HHS","NCRR NIH HHS","NHLBI NIH HHS","National Natural Science Foundation of China","National Cancer Institute","NCI NIH HHS","National Heart Lung and Blood Institute","CSRD VA"],"pagination":["218134"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12640559"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["636"],"pubmed_abstract":["The recent advancements in computational pathology focus on extracting valuable prognostic insights from whole-slide images (WSIs). These methods primarily involve deep learning-based or handcrafted feature representations of the disease's morphologic patterns associated with outcomes. However, determining the most prognostic regions within tumors remains challenging due to significant morphologic heterogeneity even within manually annotated tumor areas. In other words, the question is not simply what type of representation is appropriate to predict cancer outcomes, but specifically where to mine those representations. To address this issue, a deep learning framework to identify prognostically relevant (PR) cancer regions (DOVER) within WSIs is presented. DOVER leverages patterns mined from the tissue microarray (TMA) spots with the associated long-term clinical outcomes. The prognostic patterns learned from the individual spots of the TMA (morphologically consistent) are then mapped into larger WSIs to locate PR regions for subsequent feature representation and patient outcome prediction. DOVER improves prognostic prediction in terms of c-index over 20 % (p < 0.05) across 2041 patients (NSCLC: n = 1141; OPSCC: n = 900). Moreover, correlations with quantitative immunofluorescent (QIF) images reveal a diverse CD8<sup>+</sup>, CD20<sup>+</sup>, CD4<sup>+</sup>, and tumor cell distribution in DOVER-selected regions, reflecting a complex interplay between tumor and immune cells. DOVER identifies statistically significant differences between PR regions, both at the molecular and morphological levels. DOVER could help identify specific spatial locations on WSIs that could be used to mine prognostic feature representation for subsequent predictions of clinical outcomes. With additional validation, DOVER could also potentially help to guide AI-informed molecular profiling of tumors."],"journal":["Cancer letters"],"pubmed_title":["A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images."],"pmcid":["PMC12640559"],"funding_grant_id":["R01 HL151277","R01 CA268287","UL1 TR002548","U54 CA254566","C06 RR012463","R43 EB028736","R01 CA216579","U01 CA248226","R01 CA257612","R01 HL158071","I01 CX002622","I01 CX002776","R01 CA249992","U01 CA113913","I01 BX004121","IK6 BX006185","P30 CA138292","U01 CA239055","R01 CA220581","R01 CA202752","R01 CA268207","U01 CA269181","R01 CA208236"],"pubmed_authors":["Koyfman S","Barrera C","Lewis J","Yang K","Song B","Scahlper K","Lu C","Chen Y","Wang X","Madabhushi A","Zhou Y"],"additional_accession":[]},"is_claimable":false,"name":"A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images.","description":"The recent advancements in computational pathology focus on extracting valuable prognostic insights from whole-slide images (WSIs). These methods primarily involve deep learning-based or handcrafted feature representations of the disease's morphologic patterns associated with outcomes. However, determining the most prognostic regions within tumors remains challenging due to significant morphologic heterogeneity even within manually annotated tumor areas. In other words, the question is not simply what type of representation is appropriate to predict cancer outcomes, but specifically where to mine those representations. To address this issue, a deep learning framework to identify prognostically relevant (PR) cancer regions (DOVER) within WSIs is presented. DOVER leverages patterns mined from the tissue microarray (TMA) spots with the associated long-term clinical outcomes. The prognostic patterns learned from the individual spots of the TMA (morphologically consistent) are then mapped into larger WSIs to locate PR regions for subsequent feature representation and patient outcome prediction. DOVER improves prognostic prediction in terms of c-index over 20 % (p < 0.05) across 2041 patients (NSCLC: n = 1141; OPSCC: n = 900). Moreover, correlations with quantitative immunofluorescent (QIF) images reveal a diverse CD8<sup>+</sup>, CD20<sup>+</sup>, CD4<sup>+</sup>, and tumor cell distribution in DOVER-selected regions, reflecting a complex interplay between tumor and immune cells. DOVER identifies statistically significant differences between PR regions, both at the molecular and morphological levels. DOVER could help identify specific spatial locations on WSIs that could be used to mine prognostic feature representation for subsequent predictions of clinical outcomes. With additional validation, DOVER could also potentially help to guide AI-informed molecular profiling of tumors.","dates":{"release":"2026-01-01T00:00:00Z","publication":"2026 Jan","modification":"2026-06-06T13:29:08.128Z","creation":"2026-05-31T03:07:29.474Z"},"accession":"S-EPMC12640559","cross_references":{"pubmed":["41238099"],"doi":["10.1016/j.canlet.2025.218134"]}}