Re: “Oncolytic Measles Virus Encoding Interleukin-12 Mediated Antitumor Activity and Immunologic Control of Colon Cancer In Vivo and Ex Vivo ” by Wang et al.

I would like to comment on the article “Oncolytic Measles Virus Encoding Interleukin-12 Mediated Antitumor Activity and Immunologic Control of Colon Cancer In Vivo and Ex Vivo” published recently in Cancer Biotherapy and Radiopharmaceuticals. ¹ This article caught my interest since my group originally published the viral vector MeVac FmIL-12², which this article deals with.

I would like to indicate several points in the article that are unclear or inconsistent. Several experiments cannot have been conducted as described.

First, the source of the viral vector and its production should be described in more detail. The statement “The cassette encoding IL-12 fusion protein […] was generated based on results from Lieschke et al.” (see Veinalde et al. ² ) describes the transgene cassette. However, information on the viral vector backbone and insertion of the transgene cassette is lacking. Generation of negative-strand RNA viruses such as recombinant measles viruses is usually performed using a reverse genetics system. ³

The authors state that cells were “transfected with MeVac and MeVac FmIL-12.” To generate the negative-sense RNA viral vector, antigenomic cDNA plasmids are co-transfected with plasmids encoding the RNP components. However, functional studies are performed by infecting cells at appropriate multiplicities of infection. The species origin of the transgene does not match the downstream experiments. “FmIL-12” designates a murine interleukin (IL)-12, but the main functional experiments in the article were carried out with human cell lines and rats.

Figure 1 displays expression of the IL-12 transgenes after “transfection” of human colorectal cancer cells and secretion of interferon-γ after stimulation of splenocytes with cell culture supernatants. ¹ Interestingly, although working with a different cell line, the expression kinetics are very similar to our previous results (Veinalde et al., ² figure 1C and D). Typically, murine cells are less permissive for measles virus compared with human cells.

Figures 2 and 3 show IL-12 expression, downstream cytokine expression and apoptosis in human cancer cell lines. ¹ These inflammatory cytokines are typically expressed by immune cells upon exposure to IL-12. Epithelial cancer cells such as the Caco-2 and HCT116 cell lines used here do not express the IL-12 receptor (see e.g., the Human Protein Atlas, https://www.proteinatlas.org/ENSG00000096996-IL12RB1, https://www.proteinatlas.org/ENSG00000081985-IL12RB2) and thus are irresponsive to IL-12. Furthermore, in the experiments reported, “negative control” refers to a viral vector not containing an additional transgene. However, this vector typically induces a similar rate of cell death as the IL-12 vector in cancer cell lines (see Veinalde et al., ² supplementary figure S2 as well as the follow-up study Backhaus et al., ⁴ figure 3). The reported difference would prompt further experimentation to decipher possible underlying mechanisms. Moreover, it is rather surprising that the two cell lines (Caco-2 in Fig. 2 and HCT116 in Fig. 3) yield almost identical results in these assays. We have previously compared several human colorectal cancer cell lines and found significant differences in their permissiveness for oncolytic measles virus (Veinalde et al., unpublished data).

Figure 4 reports in vivo experiments. Sprague-Dawley rats were used for these studies. ¹ This is an outbred immunocompetent strain. However, apparently human cancer cells (CaCo-2) were used as tumor model that typically engraft only in immunodeficient animals. Thus, this animal experiment cannot have been carried out as described in the publication.

As with the in vitro apoptosis assay, also in vivo “mock” (referring here to a vector without an immunomodulatory transgene) is expected to elicit oncolytic and also immunotherapeutic effects, which are enhanced by additional IL-12 expression (see Veinalde et al., ² figure 2 and Backhaus et al., ⁴ figure 5). Consequently, some degree of inflammatory cytokine expression would be expected also in the “mock” group (see Veinalde et al., ² figure 3).

Perhaps these contradictions can somehow be resolved; otherwise, the validity of the publication must be questioned.