PXD073928

PXD073928 is an original dataset announced via ProteomeXchange.

Title	AI-based digital pathology and spatial proteomics enable precision oncology: A case report of recurrent melanoma in a young patient
Description	Background: Metastatic melanoma presents clinical challenges due to tumor heterogeneity and resistance mechanisms, especially when standard treatments fail. Recent advances in artificial intelligence (AI) and spatial proteomics have opened new avenues for precision oncology. Exploring these principles, we report the development and clinical application of a novel integrative workflow combining AI-based digital pathology with spatial proteomic profiling to support personalized treatment strategies. Methods: High-resolution H&E images were used to create and validate an AI model able to uncover subtle morphological patterns and identify different subpopulations of melanoma cells and stroma components. Laser microdissection was then used to isolate clinically relevant regions, and MS-based proteomics was applied to map the spatial proteome signature across different regions. This workflow was applied to a challenging clinical case involving a young patient with recurrent melanoma presenting with multiplex metastases. In addition to the primary tumor, lung and brain metastases were analyzed. Results: Our AI method unveiled two distinct cell subpopulations (PT1 and PT2) spatially separated within the primary lesion. Spatial proteomics further revealed inter-tumor heterogeneity between the primary tumor and metastatic lesions, with increased kinases correlated to target drug resistance. Additionally, upregulated glycolytic signaling and mitochondrial metabolism were identified as key drivers of melanoma progression in this patient. Notably, the AI model also pinpointed PT1 as the likely driver of metastasis, based on strong morphological similarities to the cells found in both brain and lung lesions. This was further supported by spatial proteomic data, which showed a close molecular match between PT1 and the metastatic sites. Conclusions: The findings suggest that targeted therapies may provide limited benefit on their own, while the combination with metabolic inhibitors could represent a more effective treatment option for the future. This case sets the ground for omics analysis inside clinical surroundings, showing how integrating AI-based pathology with spatial proteomics can uncover metastatic drivers and support personalized cancer care in the real world.
HostingRepository	PRIDE
AnnounceDate	2026-09-07
AnnouncementXML	Submission_2026-09-07_04:20:10.865.xml
DigitalObjectIdentifier
ReviewLevel	Peer-reviewed dataset
DatasetOrigin	Original dataset
RepositorySupport	Unsupported dataset by repository
PrimarySubmitter	Jeovanis Gil
SpeciesList	scientific name: Homo sapiens (Human); NCBI TaxID: NEWT:9606;
ModificationList	iodoacetamide derivatized residue
Instrument	Q Exactive HF

Revision	Datetime	Status	ChangeLog Entry
0	2026-02-02 06:32:05	ID requested
⏵ 1	2026-09-07 04:20:11	announced

Guedes J, Woldmar N, Baltas E, Kriston A, Migh E, Kov, á, cs F, Oskol, á, s H, Marko-Varga M, Appelqvist R, Wieslander E, Szasz AM, Ver, é, b Z, Gyulai R, Bal, á, zs N, é, meth I, Malm J, Horvath P, Gil J, Marko-Varga G, AI-based digital pathology and spatial proteomics enable precision oncology: A case report of recurrent melanoma in a young patient. NPJ Precis Oncol, 10(1):(2026) [pubmed]

10.1038/s41698-026-01569-w;

submitter keyword: Precision Oncology, AI-Digital pathology, laser capture microdissection, melanoma, spatial proteomics

Jeovanis Gil
contact affiliation	Department of Translational Medicine, Lund University, Sweden
contact email	jeovanis.gil_valdes@med.lu.se
lab head
Jeovanis Gil
contact affiliation	Lund University
contact email	jeovanis.gil_valdes@med.lu.se
dataset submitter

Dataset FTP location NOTE: Most web browsers have now discontinued native support for FTP access within the browser window. But you can usually install another FTP app (we recommend FileZilla) and configure your browser to launch the external application when you click on this FTP link. Or otherwise, launch an app that supports FTP (like FileZilla) and use this address: ftp://ftp.pride.ebi.ac.uk/pride/data/archive/2026/09/PXD073928

PRIDE project URI

• PRIDE
  1. PXD073928
     1. Label: PRIDE project
     2. Name: AI-based digital pathology and spatial proteomics enable precision oncology: A case report of recurrent melanoma in a young patient