The MYC oncogene is commonly activated in hepatocellular carcinoma. In a letter to Nature, Dean Felsher and colleagues describe how they have used a conditional transgenic mouse model to study expression of MYC in liver cells and show that MYC inactivation induces tumour regression and dormancy.

To establish a suitable model, the authors crossed a mouse in which the liver activator protein (LAP) promoter drives expression of the tetracycline transactivating protein (tTA) in liver cells, with a mouse in which MYC is under control of the tetracycline response element (which is bound by tTA) — producing the LAP-tTA/tet-off-MYC mouse. Only mice expressing both the Lap and Myc transgenes and not treated with doxycycline — which also regulates tTA — expressed MYC and developed hepatocellular carcinomas.

As liver tumours are usually refractory to therapy, the authors expected that inactivation of MYC in the mice with established liver tumours would be ineffective in causing tumour regression. However, this was not the case, 50 transgenic mice moribund with liver tumours showed rapid and sustained tumour regression when treated with doxycycline to inactivate MYC.

So, what mechanism induces this regression? Within a few days of treatment with doxycycline, the tumour cells had lost their high mitotic index and markers of high proliferation, and showed increased apoptosis. In addition, expression of the immature differentiation marker α-fetaprotein decreased and expression of the hepatocyte marker carcinoembryonic antigen increased — the cells had differentiated into normal liver cells. When tumour cells were transplanted into the skin of severe combined immunodeficient (SCID) mice and MYC was inactivated, tumours quickly regressed and normal liver cells resembling hepatic lobules were observed. Interestingly, reactivation of MYC expression led to tumour regrowth with identical histology and genomic signatures to the original tumour, and the authors therefore conclude that the tumour cells remain dormant and retain the capacity to regain neoplastic features.

If this hypothesis is true, some tumour cells must persist after MYC inactivation, so the authors crossed the LAP-tTA/tet-off-MYC mice with mice transgenic for luciferase, so that the liver tumour cells transplanted into SCID mice could be tracked by measuring light emitted by luciferase activity. Eight months after MYC inactivation, luciferase activity was still detectable in tumour cells, whereas normal liver cells were undetectable 5 days after MYC inactivation.

Serial transplantation of liver tumours with inactivated MYC only rarely led to relapse — the relapsed tumours had compensatory increases in L-MYC and N-MYC. The authors conclude that targeted inactivation of MYC might be an effective treatment for some liver cancers.