Abstract
Cancer-associated fibroblasts (CAF) potentiate the response of cancer cells to oncolytic viruses.
Major finding: Cancer-associated fibroblasts (CAF) potentiate the response of cancer cells to oncolytic viruses.
Mechanism: CAFs are more sensitive to infection and secrete FGF2 to suppress antiviral responses in cancer cells.
Impact: Oncolytic viruses expressing FGF2 exhibit increased therapeutic efficacy in tumor-bearing mice.
Oncolytic viruses (OV) are currently in clinical trials as anticancer drugs, and an improved understanding of how tumor cells respond to OVs may reveal useful information for the design of more effective therapeutics. Paracrine interactions between cancer cells and cancer-associated fibroblasts (CAF) have been shown to modulate the response to several anticancer therapeutics, prompting Ilkow and colleagues to investigate whether CAFs also affect the response of cancer cells to OV therapy. TGFβ1- or cancer cell–induced CAFs were significantly more susceptible to OV infection compared with normal fibroblasts due to reduced baseline expression of genes involved in type I IFN signaling and antiviral responses, including retinoic acid–inducible gene 1 (RIG1, also known as RARRES3), relative to normal fibroblasts. Coculture with CAFs rendered various cancer cell lines more susceptible to OV infection, suggesting that paracrine factors secreted by CAFs mediate virus sensitization of cancer cells. Of the cytokines and growth factors known to be present in the tumor milieu, only FGF2, which was upregulated in CAFs, significantly enhanced OV replication in cancer cells. Cancer cells cocultured with CAFs or exposed to FGF2 exhibited reduced expression of RIG1 and downstream antiviral proteins. In vivo, intratumoral injection of FGF2 led to increased OV replication in several xenograft models; similar virus sensitization was observed in response to coinjection of pancreatic fibroblasts. Furthermore, analysis of human pancreatic cancer samples revealed that high tumor FGF2 expression correlated with increased susceptibility to OV infection ex vivo, suggesting that FGF2 may be a potential biomarker. Importantly, an OV engineered to express an FGF2 transgene more effectively decreased tumor burden compared with the parental virus and without toxicity to normal cells. These results identify reciprocal cross-talk between cancer cells and CAFs in the tumor microenvironment as a determinant of OV sensitivity and support further development of FGF2-expressing OV therapeutics.