By Declan Butler of Nature magazine

How much radiation is 'unsafe' for humans? For those exposed to fallout from the disaster at Japan's Fukushima Daiichi nuclear plant, the question is all too real. But there is still no good answer: the accident has highlighted the enormous difficulties in estimating the long-term health risks of relatively low doses of radiation.

A group of leading researchers in Europe had hoped that a fresh round of studies on people exposed to radiation after the Chernobyl nuclear disaster in 1986 would finally begin to help fill this yawning science gap (see 'Lessons from the past'). But their proposal is now looking increasingly unlikely to proceed.

The Chernobyl lifespan cohort study was one of the main components of the Agenda for Research on Chernobyl Health (ARCH), which was proposed last year by an international panel of experts who had been charged by the European Commission to advise it on future research needs (see 'Chernobyl's legacy'). The study would track the lifetime health of more than half a million 'liquidators' sent in to clean up the area around Chernobyl, as well as of the general population of the region who were children at the time of the accident. The power of the study would lie in its size, offering more than ten times as many people as the lifetime cohort study set up in Japan after the Hiroshima and Nagasaki atomic bombs, which remains the gold standard for studies on the impact of radiation on a population.

Such lifetime cohort studies have never been established for Chernobyl, points out ARCH panel member Dillwyn Williams, a cancer researcher at the Strangeways Research Laboratory in Cambridge, UK, adding that this is probably the last opportunity to set them up. "If ARCH is not supported they probably never will be," he says. The study promises a new way to study long-term health effects - including cancers, but also many other diseases -- following a nuclear accident. Unlike the survivors of the atomic bombs in Japan, the Chernobyl cohorts were exposed to a wide range of radiation doses over a long period of time, making them much more relevant to the Fukushima accident and potentially helping to refine protection levels for radiation workers.

Picking up the MELODI

Setting up the study would cost around €1 million (US$740,000), with annual running costs of a similar order, according to Keith Baverstock, a radiobiologist at the University of Eastern Finland in Kuopio, and a member of the ARCH panel. Many of the registries of potential cohort members already exist in Ukraine, Russia, Belarus and the Baltic countries, says Baverstock, and just need to be properly exploited. But as Williams notes, this would require a level of long-term financing that he feels only a dedicated initiative from the European Commission could provide.

The commission, however, seems not to favor this approach. In response to an enquiry from a member of the European Parliament, seen by Nature, Máire Geoghegan-Quinn, the research commissioner for the European Union (EU), wrote on 5 September that the ARCH proposal was outside the remit of the EU's Euratom nuclear energy research programs. She added that efforts to "deepen understanding of the interaction of radiation with tissue" would be tackled under a new European platform: the Multidisciplinary European Low Dose Risk Research Initiative (MELODI).

Led by 15 nuclear research centers and national radiation protection agencies, MELODI aims to integrate their programs and coordinate research and funding across the continent to study the effects of radiation from environmental sources, medical diagnostics, or other sources.

But some scientists, including Baverstock, say that this would effectively kill off the ARCH proposal. MELODI "is not organized in such a way that it could set up and provide long-term support for cohort studies of the type we need," says Williams. MELODI's main focus is on experimental work to study the mechanisms of radiation damage, and it plans to do little of the messier epidemiological work that is needed to gather data and to get a picture of any health effects in the aftermath of a real-life nuclear accident. Many experts argue that both mechanistic and epidemiological approaches are essential to understand the implications of radiation exposure (see 'We don't know enough about low-dose radiation risk').

Uncertain doses

MELODI's strategy is currently being thrashed out by a six-year initiative, the Low Dose Research towards Multidisciplinary Integration (DoReMi) project, set up last year. Most of its €13 million in funding will go towards research, infrastructure and procedures that will lay the groundwork for the structure of MELODI.

Sisko Salomaa, a co-founder of MELODI, coordinator of DoReMi, and research director of the Finnish Radiation and Nuclear Safety Authority, acknowledges that MELODI and ARCH are quite different, and that it would be difficult to accommodate the Chernobyl lifespan study within MELODI. Responsibility for any further studies of the Chernobyl cohorts should mainly lie with the affected states, she also argues, with any European involvement a matter for discussion. Salomaa also notes that uncertainties over the doses received by the exposed Chernobyl population potentially weaken the value of such studies.

According to the commission, some aspects of the ARCH proposal will go ahead as part of two other projects called PROCARDIO and CEREBRAD. Beginning on 1 October, these efforts will assess the risk of cardiac, cerebral and vascular diseases from low doses of radiation. MELODI has slated a discussion of the ARCH proposal at a session during its next workshop in Rome in November.

Williams, however, is adamant that the cohort study should proceed under a dedicated funding stream. Referring to the cost of creating a new shelter for the highly radioactive remnants of the Chernobyl disaster (see 'Cash promised to help clean up Chernobyl'), he argues that "if the Commission can find half a billion euros to put a roof on the reactor, I think they should be able to find money for long-term funding to study the effects of the accident."

This article is reproduced with permission from the magazine Nature. The article was first published on September 30, 2011.