Introduction:

IFNα gene therapy is rapidly evolving. With a notable response rate of 31% in a phase 3 trial, IFNα gene therapy is under consideration for approval for patients with BCG Unresponsive Non-Muscle Invasive Bladder Cancer (NMIBC). Understanding and targeting mechanisms of resistance to this therapy is of outmost importance. Our preclinical studies demonstrated that IFNα alters tumor cell metabolism and subsequently affects treatment resistance. Using cell lines, preclinical models, and clinical samples, we investigated the role of IFNα in regulating tumor metabolic pathways.

Methods:

Murine bladder cancer (BLCA) cell lines were treated with recombinant IFN (rIFNα) or Lentiviral IFNα (LV-IFNα) and sequencing was performed to identify differentially altered genes and pathways. Subcutaneous and intravesical tumor models were used to assess the effect of IFNα on tumor metabolism. Lipidomics, glucose uptake, and lactate levels were measured following treatment with IFNα in murine models. Lipidomics analysis was performed on phase 2 Nadofaragene firadenovec urine samples, collected at day 1 (pre-treatment) and day 12 (post-treatment). For this, we selected 7 responders and 6 non-responders to Nadofaragene firadenovec.

Results:

Results

Several genes involved in fatty acid anabolism were downregulated and several genes involved in glycolysis were upregulated following treatment with IFNα in cell lines, in mouse models, and in human tumors treated with Nadofaragene. Q-PCR was used to confirm changes in these genes. We performed lipidomics on mouse tumors treated with poly(I:C), an IFNα surrogate, and identified 79 lipids that were upregulated and 12 that were downregulated. Three major classes of lipids were upregulated: phosphatidyl choline, sphingomyelin and phosphatidyl ethanolamine; and one major class (cardiolipin) was downregulated, Figure 1.

On lipidomics analysis on urine samples, 592 lipids were detected in the urine samples tested. The lipid profile was distinct between responders and non-responders before treatment and the lipid profile is altered in non-responders after treatment. Finally, we found increased glucose uptake, and lactate levels suggesting increased glycolysis in BLCA cells treated with IFNα, Figure 2.

Conclusion:

We found that bladder cancer tumor cells tend to present a metabolic shift toward a fatty acid-based metabolism. Response to IFNα gene therapy is associated with a metabolic shift towards a glycolytic-based metabolism. We uncovered a novel mechanism whereby IFNα is able to modulate tumor cell metabolism and may contribute to response to it.  Combining IFNα-based therapy with approaches targeting tumor cell metabolism might further facilitate in overcoming resistance to IFNα gene therapy and ultimately improve patients' outcomes.

Funding: Research is supported in part by A.I. Virtanen Institute for Molecular Sciences (Kuopio, Finland) and MD Anderson CCSG program (P30 016672)