Researchers at NYU School of Medicine have found that Sindbis virus, a Togavirus that causes cold-like symptoms in humans, can systemically and specifically target tumour cells in mice. The press release is here, and the original paper in Nature Biotech is here. The virus enters mammalian cells by binding to the 67kDa high-affinity laminin receptor (LAMR), which makes it highly selective for tumour cells. Relative to healthy cells, the vast majority of tumour cells express greatly increased numbers of LAMR on their surfaces, and (again, unlike healthy cells) the majority of those receptors are unoccupied. (Normal LAMR function is involved in interaction with the extracellular matrix.)

Tseng and colleagues showed that Sindbis virus could infect tumours without infecting surrounding normal tissue and cause significant reduction in tumour mass (up to complete regression in some models) whether they induced the tumours subcutaneously or in the pancreas, lungs or peritoneum of immunocompromised mice using human cancer cells, or subcutaneously in immunocompetent mice using mouse cancer cells. This indicates that the results are not species specific and that a working immune system does not interfere with the anti-tumour activity despite repeated treatments with the virus. They also showed that the virus could infect and reduce spontaneous tumours in a cancer-prone mouse breed, decreasing the likelihood that the anti-tumour activity depended in some way on the manner in which the tumours were induced.

The virus was injected intraperitoneally or intravenously at as great a remove from the tumor sites as possible, which means that the virus targeted the tumours after being disseminated in the blood. This is great news, because it indicates that we can build a tumour-seeking virus missile that will find metastatic tumours no pathologist or surgeon could detect. It's also important to note that Sindbis is not a retrovirus (and so does not integrate its genome into the host cell's) and is highly lethal, so it kills virtually any cell it enters well before that cell could become a Sindbis-induced cancer itself. The virus used in this study was created in such a way that it cannot package itself into new particles after infecting a cell: it is replication incompetent, and cannot spread on its own through the patient's body. (Barring frankenviral recombination events, it only gives you a cold anyway.)

The images are from a different paper entirely (Zhang et al. J Virol vol. 76 pp. 11645-11658). Right, cryoelectron microscopy map of a partially deglycosylated Sindbis particle; left, surface topology model of same. Both pictures are to the same scale, the bar in the map represents 20 nm.