Digital spatial profiling of advanced solid tumors and lymphomas from a phase 1 trial of copanlisib and nivolumab.
3127 Background: Digital spatial profiling (DSP) is an innovative technique that facilitates spatially resolved proteotranscriptomic analysis within tissue sections, providing essential insights into the tumor microenvironment (TME). As part of the exploratory objectives in the Phase 1B trial ( NCT03502733 ) evaluating adult patients(pts) with solid tumors and lymphomas copanisib (C), nivolumab (N) + ipilumumab (I), we employed the NanoString-Bruker GeoMx DSP platform to evaluate spatial heterogeneity in differentially regulated biomarkers. Methods: The study analyzed samples from pts in the trial's doublet treatment arm, which included C + N. Pts receive C on days 1 and 15 or days 1, 8, and 15 of each cycle and nivolumab on day 1 or days 1 and 15 of each cycle. Core biopsies collected from 8 pts at cycle 1 day 1 pre-dose (C1D1, baseline), cycle 1 day 8 post-dose (C1D8, C only) and cycle 2 day 15 post-dose (C2D15, C+N) were selected for GeoMx analysis. Tissue sections (5-μm) from archival FFPE blocks were prepared on glass slides and hybridized with photocleavable tag-conjugated antibodies (targeting 85 proteins) and oligonucleotide probes (whole transcriptome- WTA) for protein and WTA analyses respectively. Tissue imaging was performed via high-resolution fluorescent microscopy using morphology markers (cytokeratin AE1/AE3, CD45, CD3, CD20, Syto-13 nuclear marker). At least three rectangular (660 x 784 μm) regions of interest (ROI) per timepoint were analyzed using NanoString nCounter for proteomics and NGS for WTA. Data quality control and analysis were conducted using the GeoMx DSP Control Center (V-3.0) with a significance threshold of α = 0.05. Results: In two lymphoma pts with stable disease (SD) or partial response (PR), PI3K downstream signaling showed downregulation at C1D8 due to C-mediated PI3K-AKT signaling inhibition, possibly through PIK3IP1 overexpression. This signaling returned to baseline by C2D15, likely due to C’s elimination half-life. In a follicular lymphoma case (#18, PR), FOXP3 expression decreased at both C1D8 and C2D15, while CD4 and CD8 levels remained constant. Immune marker expression (PD-L1, PD-1, CTLA-4, CD80) progressively declined. However, in diffuse large B-cell lymphoma (#14, SD), no changes in T-cell markers were observed. Solid tumor cases (#12, #24, SD) showed PI3K-AKT signaling downregulation at C2D15, along with CD3+/CD8+ T-cell infiltration into the tumor. FOXP3 levels slightly decreased in tumor and immune compartments. Progressive disease cases showed no change in T-cell markers. Conclusions: GeoMx DSP demonstrated its capability to investigate phospho-signaling and immune profiles in tumor and stromal compartments of small biopsies, highlighting its potential to enhance the understanding of TMEs in clinical studies. Further applications may provide critical insights for clinical cancer trials. Clinical trial information: NCT03502733 .
Authors
- Geraldine O’Sullivan Coyne (ORCID: https://orcid.org/0000-0002-2918-6326)
- Chris Karlovich (ORCID: https://orcid.org/0000-0002-4617-2161)
- Barry C. Johnson (ORCID: https://orcid.org/0000-0001-9482-2248)
- Parimal Kumar (ORCID: https://orcid.org/0000-0003-1982-7983)
- Peter Wu (ORCID: https://orcid.org/0000-0001-8137-9179)
- Alice P. Chen (ORCID: https://orcid.org/0000-0002-1426-853X)
- Christina Rosenberger
- Lindsay Dutko
- Jibran Ahmed (ORCID: https://orcid.org/0000-0002-3412-7689)
- Sayak Ghatak (ORCID: https://orcid.org/0000-0002-3431-9955)
- Biswajit Das (ORCID: https://orcid.org/0000-0001-7225-2926)
- Katharina Paulus
- Li Chen (ORCID: https://orcid.org/0000-0002-3965-8833)
- Mary Jane Ong
Institutions
- Leidos (United States) (US)
- National Institutes of Health (US)
- Frederick National Laboratory for Cancer Research (US)
- National Cancer Institute (US)
- Leidos Biomedical Research Inc. (United States)
Publication Details
- Journal
- Journal of Clinical Oncology
- Published
- 2025-05-28
- DOI
- https://doi.org/10.1200/jco.2025.43.16_suppl.3127
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00