Friday, August 7, 2009 - 9:20 AM

COS 117-5: Transmission dynamics of Tasmanian devil facial tumor disease: Implications for developing, modelling, and testing control strategies

Hamish I. McCallum1, Menna E. Jones1, Shelly Lachish2, and Nick Beeton1. (1) University of Tasmania, (2) University of Queensland

Background/Question/Methods

The largest surviving marsupial carnivore, the Tasmanian devil Sarcophilus harrisii is threatened with extinction by an infectious cancer, Tasmanian devil facial tumor disease. The tumor is an infectious cell line, thought to be spread by biting, which is able to transmit between individuals because of very low genetic diversity. One of the few management strategies available for infectious disease in an endangered species is disease suppression through removal of infected animals. Understanding the transmission dynamics, especially estimating the basic reproductive rate R0 and determining whether transmission is dependent on host density, is essential to evaluate whether this approach is feasible.  Using mark- recapture data, we have estimated the rate of increase in prevalence through time to derive estimates of R0 and have in turn used these estimates to model the rate of removal of infected animals necessary to successfully suppress disease. Since 2006, we have trialled this control strategy on the semi-isolated Forestier Peninsula, using multistate mark recapture methods to estimate the rate of transition from healthy to infected classes.

Results/Conclusions

Recapture probabilities within trapping bouts did not depend on infection status, providing confidence that prevalence in the trapped population was an unbiased estimate of disease prevalence.  The rate of increase in prevalence at sites monitored since disease arrival was not clearly associated with local population size. High levels of prevalence were maintained despite decreases in population density of up to 90%. This information suggests that transmission is frequency rather than density dependent. Estimation of R0 was hampered by poor knowledge of the incubation period of the disease. However, our models suggest that the rate of removal of infected animals that is necessary to suppress disease would be difficult to achieve in the field. On the Forestier Peninsula, where the trial is being undertaken, there is as yet no evidence that the transition rate from healthy to infected classes has been reduced, relative to the transition rate observed at an unmanipulated site; nor is the rate of decline in the devil population slower at the manipulated site. We conclude that the current approach has not been sufficient to suppress the disease on the Forestier Peninsula. If it cannot work in an almost closed population, it is even less likely to be successful on a larger scale in open populations. We are currently exploring alternative disease suppression approaches.