Establishing a three-dimensional organoid model of primary breast cancer for assessing effects of oncolytic virotherapy

Abstract

Although several oncolytic viruses have already been tested in early-stage clinical studies of breast cancer, there is still an urgent need to develop patient-derived experimental systems that mimic the response of breast cancer to oncolytic agents in preparation of testing different oncolytic viruses in clinical trials. We addressed this need by developing a protocol to study the effects of oncolytic viruses in stable organoid cell cultures derived from breast cancer tissue and control tissue.

We used an established three-dimensional organoid model derived from tissue of 10 patients with primary breast cancer. Furthermore, we established an organoid model derived from healthy control tissue from 6 patients. We developed an experimental protocol for infecting organoid cultures with oncolytic viruses and compared the oncolytic effects of a measles vaccine virus (MeV) and a vaccinia virus (GLV) genetically engineered to express either green fluorescent protein (MeV-GFP) and red fluorescent protein (GLV-0b347), respectively, or a suicide gene encoding a fusion of cytosine deaminase with uracil phosphoribosyltransferase (MeV-SCD and GLV-1h94, respectively), thereby enabling enzymatic conversion of the prodrug 5-fluorocytosine (5-FC) into cytotoxic compounds 5-fluorouracil (5-FU) and 5-fluorouridine monophosphate (5-FUMP).

The method demonstrated that all four oncolytic viruses significantly inhibited cell viability in organoid cultures derived from breast cancer tissue. The oncolytic effects of the oncolytic viruses expressing suicide genes (MeV-SCD and GLV-1h94) were further enhanced by virus-triggered conversion of the prodrug 5-FC to toxic 5-FU and toxic 5-FUMP.

The model therefore provides a promising in vitro method to help further testing and engineering of new generations of virotherapeutic vectors for in vivo use.