Page 64 - SRPSKO DRUŠTVO ISTRAŽIVAČA RAKA

Serbian Association for Cancer Research SDIRSACR

unresolved issue is how aggressive cancer cells circumvent the initial suppressive nature of fibroblasts to ultimately
convert them into tumor-supporting CAFs. The identity of the cancer-derived factors that mediate this reprogramming
is still largely unknown. Our own work has shown that metastatic breast cancer cells, unlike their non-aggressive
counterparts, secrete IL-1β, which binds to the IL-1 receptor on CAFs and suppresses the secretion of asporin (ASPN)
(Maris et al. PLOS Med. 2015). ASPN is a potent inhibitor of TGF-β1, a key driver of epithelial-to-mesenchymal transition
(EMT) and stemness in cancer cells. Similarly, Ijichi et al. reported that pancreatic cancer cell-derived CXCL1 induces
CTGF expression in CAFs to promote tumor growth (Ijichi et al. J Clin Invest. 2011). In yet another recent study, our group
has reported in hepatocellular carcinoma that tumor-suppressive CAF can secrete prolargin (PRELP) which can bind and
inhibit a set of key growth factors that promote tumor progression (Chiavarina et al. Oncogene 2022). These findings,
while illuminating, represent only a fraction of the complex cancer–stroma interactions, and many more remain to be
discovered—particularly in the context of specific tumor types and therapy resistance. The failure of indiscriminate
CAF targeting using Sonic hedgehog (SHH)-smoothened (SMO) signaling inhibitors in clinical trials (Catenacci et al. JCO
2015) underscores the need for a more nuanced understanding of CAF function. Collectively, these insights call for a
paradigm shift in how we study and therapeutically exploit the tumor stroma, especially under the selective pressures
imposed by cancer treatments. To this end, our lab has recently embarked on a combination of spatial and single-cell
omics in multiple cancers including liver metastases. This approach permits for the first time a better understanding of
CAF tumor heterogeneity, offering additional means to differentiate tumor promoting versus tumor suppressing CAF-
derived molecules (Giguelay et al. Theranostics 2022, Honda et al. Theranostics 2024).

Acknowledgments and funding: French National Research Agency (ANR), National Cancer Institute (INCA) and
European Union Horizon Europe.

L08
Understanding Factors Influencing Immunotherapy Response in Head and Neck Cancer

Housaiyin Li 1,2,3 , Dan Zandberg 1,4,5 , Aditi Kulkarni , Simion I. Chiosea , Brian R. Isett , Gabriel L. Sica , Riyue Bao 1,2,4 ,
1,4
1,7
1,6
1,7
Jing H. Wang 1,4,5,8 , Robert L. Ferris and Lazar Vujanovic 1,2,5,6
9
¹UPMC Hillman Cancer Center, Pittsburgh, PA, USA
2Tumor Microenvironment Center, UPMC Hillman Cancer Center, Pittsburgh, PA, USA
3Molecular Genetics and Development Biology Graduate Program, University of Pittsburgh, Pittsburgh, PA, USA
4Department of Medicine, University of Pittsburgh, Pittsburgh, PA, USA
5Cancer Immunology and Immunotherapy Program, UPMC Hillman Cancer Center, Pittsburgh, PA, USA
6Department of Otolaryngology, University of Pittsburgh, Pittsburgh, PA, USA
7Department of Pathology, University of Pittsburgh, Pittsburgh, PA, USA
8Department of Immunology, University of Pittsburgh, Pittsburgh, PA, USA
9UNC Lineberger Comprehensive Cancer Center, UNC Health Care System, Chapel Hill, NC, USA

Keywords: head and neck cancer; CD8+ T cells; clinical trial; single-cell genomics; T cell dynamics; combination
immunotherapy

Background: In the evolving landscape of immuno-oncology, a key unresolved question is whether different immune
checkpoint inhibitors (ICI) and their combinations promote similar immune mechanisms during a favorable treatment
response, or whether they act through distinct pathways. To address this critical gap in knowledge, we leveraged a
novel phase II neoadjuvant clinical trial (NCT04080804) designed to enhance adaptive anti-tumor immunity in patients
with resectable, locally advanced head and neck squamous cell carcinoma (HNSCC). Patients were treated with anti-
PD-1 (nivolumab; Nivo) monotherapy, or in combination with either anti-LAG-3 (relatlimab; Nivo+Rela), a regimen
previously untested in this disease, or with anti-CTLA-4 (ipilimumab; Nivo+Ipi) monoclonal antibodies.
Materials and Methods: Forty-one patients were randomized across the three treatment arms. To evaluate treatment-
induced immune mechanisms, we performed extensive single-cell analyses using pre- and post-treatment matched
tumor specimens from 35 patients. Single-cell RNA sequencing, T cell receptor (TCR) sequencing, and Cellular Indexing
of Transcriptomes and Epitopes by Sequencing (CITE-seq) were performed on 372,914 CD45⁺CD3⁺ tumor-infiltrating
lymphocytes (TIL), including 137,133 CD8⁺ TIL. Multiplex immunofluorescence imaging was used to quantify CD8⁺ T cell
infiltration and its association with pathologic response.
Results: Both Nivo+Ipi and Nivo+Rela combination therapies elicited superior pathologic responses compared to

49