Could Ross River Virus be the next Zika?

Ross River virus (RRV) is a mosquito-transmitted Alphavirus that is indigenous to Australia with clinical cases primarily coming from sub-tropical Queensland.1 Its ecological niche is complex, involving a reservoir in native animals, and it is able to employ a variety of mosquito vectors including Aedes and Culex species. Although most infections are asymptomatic, RRV disease (epidemic polyarthritis, EPA) is a febrile illness often accompanied by a maculopapular rash that may involve the palms and soles. Pauci-articular arthritis is RRV’s most bothersome manifestation that may continue for months. In Australia, it appears as a regular, seasonal, febrile disease, but it has definite epidemic potential as demonstrated in the Southwest Pacific islands in 1979–80.2,3 Although an inactivated RRV vaccine has completed phase III clinical trials, it remains unlicensed because of a lack of commitment to its use to prevent sporadic cases of persistent arthritis. Given the recent demonstration of the ability of other arboviruses (Chikungunya and Zika) to suddenly cause epidemics, one can speculate that subtle changes to RRV could cause a similar, massive global economic and social disruption spreading across continents. RRV as an Alphavirus is more closely related to Chikungunya (CHIKV) than to the Flavivirus Zika (ZIKV), which is more like dengue. There had been epidemics of polyarthritis in New South Wales in 1928 and in soldiers stationed in northern Queensland during the Second World War. Doherty recovered a virus from Aedes vigilax mosquitoes in Townsville in 1963 and showed, on serological grounds, that it was the causative agent of epidemic polyarthritis. Almost all patients are diagnosed based on their symptoms and the detection of anti-RRV IgM antibody in acute phase serum samples. The role of RRV in the environment is complex, with a large number of vertebrates as potential hosts; man is apparently a collateral infection from the main reservoir in macropods such as kangaroos and wallabies. However, human–mosquito–human transmission may also occur. Approximately 5000 clinical cases of RRV occur each year, with RRV being the most common vector-borne disease in Australia. Although most infections are asymptomatic, it is uncertain whether the actual ratio is high (50:1) or low (1.2:1).1 In Australia, most transmission occurs during the warm months between December and April. Meteorological changes affect the risk of RRV infection, with the highest risk after excess rainfall, but this is inconsistent and models predicting RRV do not work well at the state level.4 Maps of a decade of RRV infections in Queensland have been produced, showing concentrations around Townsville and Brisbane. The estimated annual incidence was reported as 54/100 000, with a male to female ratio of 1:1.1.5 RRV disease is a dengue-like febrile illness with an incubation period of 2–21 days. Many, if not most, patients have a rash which is usually maculopapular and can involve all parts of the skin, including palms and soles. Symmetric acute arthritis often in the hand, ankle or knee is common, ranging from simple arthralgia to actual joint effusions. Most arthritic symptoms resolve over several days when treated conservatively with non-steroidal anti-inflammatory drugs.1 Some individuals may be incapacitated from joint pain for weeks to months; it is uncertain whether this is due to viral infection of the joint or an immune reaction. Whether a long-term arthritis, fatigue syndrome exists after RRV infection is problematic and possibly analogous to the situation with Lyme Disease. The current risk to travellers to Australia is very low, although sporadic cases from those visiting northern Australia at the end of the year are seen. Besides the World War Two epidemics thought to be due to RRV, a more recent 1997 outbreak occurred in US Marines, conducting exercises near Rockhampton. An estimated 1.5% infection rate was based on serological conversions among 938 Marines with an asymptomatic to symptomatic ratio of 1.2:1.6 This infection rate is the same as that for the population of Queensland, in general.2 In broad terms, Alphaviruses causing encephalitis are found in the Americas and those causing arthritis in the remainder of the world, except Antarctica. Individual Alphaviruses are thought of as being geographically restricted, but beginning in 2004 several ‘new’ strains of CHIKV spread from Africa across Asia into the Americas, causing millions of infections.7 The risk of other Alphaviruses doing the same is unknown but RRV certainly has that potential. In 1979–80, an epidemic of RRV infection spread across the Pacific affecting the islands of Fiji, Tonga, Samoa, Cook Islands and New Caledonia.1,2 The epidemic in the Pacific was found to be associated with a new strain of RRV which originated in Australia several years previously and which has now replaced all earlier strains of RRV in Australia. The epidemiology of the 1979–80 outbreaks in the Pacific provided strong evidence of human–mosquito–human transmission of RRV without other vertebrate hosts. Cases of epidemic polyarthritis began to be detected in tourists returning from Fiji in ~2000 and the first report of this occurring appeared in 2005. Subsequent reports suggest infection may also be occurring in Samoa and French Polynesia.8 A candidate RRV vaccine has completed phase III trials.9,10 It is safe in the sense that it caused few adverse events in 382 persons using a double (UV and formalin) inactivated virus preparation adjuvanted with alum. The immunization schedule used was 2.5 micrograms of protein given on days 0, 21 and 180. Efficacy determination is problematic due to the very low annual incidence of human RRV infections. In a mouse model of RRV infection, the vaccine-induced antibodies protected adult mice from viraemia. Passive immunization of mice with human vaccinee sera also was protective. Antibody-dependant enhancement of viral disease is an important problem of dengue; testing of the RRV vaccine found no evidence of such enhancement of RRV or CHIKV infection.10 Regulatory authorities have indicated what would be required to license the vaccine but in the absence of a defined, government-supported immunization programme creating a clear case for vaccine use, few companies would be prepared to bear the cost of licensing. These considerations might change quickly if transmission also began in the southern states of the USA, for example, where there are competent mosquito vectors and a population who would create a demand for the vaccine. The conundrum with the RRV vaccine is not unlike that seen with CHIKV or Zika vaccines today. Although there was an investigational vaccine available for CHIKV prior to 2000, it was never used, in part, because the absence of a market able to pay for it did not justify the cost of efficacy trials. Until the Zika epidemic occurred, this virus was regarded as a trivial novelty from Africa and Southeast Asia which would never warrant the cost of developing a vaccine. Once these viruses were causing epidemics, preparation of modern vaccines began. By the time they could be used for field efficacy testing, the epidemics had largely ceased as had the sense of urgency for their use. However, the possible teratogenic effects of ZIKV infection are providing an impetus to continue development of a Zika vaccine. Next-generation Chikungunya vaccines also are being developed but with a reduced sense of urgency. The failure of RRV, to date, to cause global outbreaks probably is due to its mosquito vectors employing a wide range of hosts, in addition to humans. Should small genetic changes, on a scale that provided epidemic potential for CHIKV, also occur in RRV, then global outbreaks of RRV infection would become a distinct possibility.9,10 Such epidemics would solve the conundrum of developing vaccines for exotic viruses such as Ross River Virus but the time required for licensure would be longer than any sudden programme to satisfy the public demand generated by an outbreak. GDS is an employee of the Australian Defence Force; no specific funding was given for this epidemiological study. The author thanks Prof John Aaskov of the Queensland University of Technology for his many discussions on arboviruses particularly Ross River virus as well as his comments on an earlier version of this manuscript. Conflict of interest: The author’s institute has previously been involved in the testing of the Ross River Virus vaccine described in the perspective, but no current staff members have any commercial interest in the vaccine’s development. G.D.S. is responsible for the entire perspective. The opinions expressed are those of the author and do not necessarily reflect those of the Australian Defence Force.