Page 63 - SRPSKO DRUŠTVO ISTRAŽIVAČA RAKA

SDIRSACR Oncology Insights

Kragujevac (projects JP15/19 and JP11/18), the Ministry of Science, Technological Development and Innovation of
the Republic of Serbia (No. 451–03–66/2024–03/200007), and from Project of Medical Faculty of Military Medical
Academy, Belgrade, Serbia, MFVMA/02/20-22, Serbia.

L07
Cancer-associated fibroblast spatial heterogeneity

Andrei Turtoi 1,2

1 Institut de Recherche enCancérologie de Montpellier, Inserm U1194 – Université Montpellier,
2 Institut du Cancer de Montpellier, Tumor Microenvironment and Resistance to Treatment Lab

Keywords: tumor-suppressing CAF, single cell RNA sequencing, therapy resistance

Metastatic dissemination is responsible for over 90% of cancer-related deaths worldwide. The liver serves as a major
hub for metastases from some of the most lethal malignancies, including colorectal, breast, lung, and pancreatic
cancers. More than 50% of patients with advanced colorectal cancer (CRC) will develop liver metastases (CRC-LM)
within five years following resection of the primary tumor. Of these patients, only one-third are eligible for surgical
intervention; the remainder must rely on systemic, chemotherapy-based treatments. Although primary tumors, such
as CRC, can often be effectively resected, metastatic disease remains largely incurable. Chemotherapy, with or without
targeted therapies, frequently leads to the development of resistance and tumor progression. These treatments lack
specificity for cancer cells, resulting in considerable toxicity to normal tissues. Consequently, chemotherapy cannot
be administered at doses sufficient to eradicate the tumor. Due to these limitations, survival for CRC-LM patients
is generally poor—median overall survival is approximately five years for operable patients and only two years for
those who are inoperable. It is now well-established that therapy-driven selection of drug-resistant clones is a key
contributor to treatment failure (Diaz et al. Nature 2012). Genomic sequencing studies have revealed profound genetic
heterogeneity among cancer cells within a single tumor (Gerlinger et al. NEJM 2012). Beyond mutational diversity, the
tumor microenvironment is a major determinant of disease progression. It is widely acknowledged that cancer cells
and the surrounding stroma form a complex ecosystem, whose collective fitness dictates natural or therapy-driven
selection (Polyak et al. Trends Genet. 2009). Furthermore, a critical yet often overlooked trait of such aggressive cancer
cells is their capacity to interact with and exploit host stromal tissue to access key resources like growth factors and
metabolites (Hunttila& De Sauvage, Nature 2013). Despite the recognized importance of the tumor microenvironment,
we still lack a comprehensive understanding of how cancer and stromal cells interact and how this interplay can be
therapeutically disrupted. Promising clinical results with immune checkpoint inhibitors (e.g., PD-L1 blockade) illustrate
that interrupting cancer–microenvironment communication can yield substantial benefits. However, only a subset of
patients currently benefits from these therapies, highlighting the need to explore other stromal components beyond
immune cells. The stromal compartment includes fibroblasts, endothelial cells, macrophages, and various immune
cells. Among them, cancer-associated fibroblasts (CAF) represent the most dynamic and multifunctional population,
implicated in virtually all hallmarks of cancer (Ronca et al. CurrOpin Oncol. 2018). CAFs are known to secrete a wide
array of growth factors, extracellular matrix proteins, immune modulators, and energy substrates. In advanced tumors,
CAFs are typically "educated" to co-evolve with cancer cells, supporting tumor progression and resistance to therapy.
For instance, hepatocyte growth factor produced by fibroblasts can rescue melanoma cells harboring mutant BRAF
from BRAF inhibition (Straussman et al. Nature 2012), while fibroblast-derived PDGF-C has been shown to counteract
anti-VEGFA therapy in murine lymphomas (Crawford et al. Cancer Cell 2009). Importantly, this pro-tumorigenic activity
of fibroblasts is not their default behavior. On the contrary, fibroblasts are naturally programmed to restrain tumor
growth (Dotto et al. PNAS 1988; Proia & Kuperwasser, Cell Cycle 2005). Emerging evidence suggests that CAFs exhibit
both tumor-promoting and tumor-suppressive roles (Chiavarina&Turtoi, Curr Med Chem 2017). Just as macrophages
can polarize into M1-like (tumor-suppressive) or M2-like (tumor-promoting) states, recent single-cell studies have
confirmed that CAFs are not a homogeneous population. Notably, these anti-tumor CAF functions appear to persist
even in advanced tumors, as demonstrated in pancreatic cancer (Özdemir et al. Cancer Cell 2014; Rhim et al. Cancer Cell
2014). Advances in single-cell technologies have led to the identification of multiple CAF subtypes in primary colorectal
and breast tumors (Li et al. Nat Genet. 2017; Costa et al. Cancer Cell 2018). In breast cancer, certain CAF subpopulations
have been directly linked to an immunosuppressive phenotype, thereby contributing to tumor progression (Costa et al.
Cancer Cell 2018). These foundational studies support the long-suspected heterogeneity of CAFs. However, whether
similar CAF subpopulations exist in metastatic lesions—and what roles they play—remains an open question. Another

48