Early Lung Cancer Detection With Spiral Computed Tomography (CT), Positron Emission Tomography (PET) and Biomarkers: Randomized Trial in High Risk Individuals

The preliminary results of the pilot study on early lung cancer detection by spiral CT and PET launched in Milan in 2000 and published in the Lancet in 2003 showed a positive outcome, confirming the essential safety of spiral CT and a high proportion (95%) of complete resections and stage I disease (77%). However, the results at six years revealed a different picture, with an increasing number of advanced lung cancers being detected from the third to the fifth year, and no evidence of significant mortality reduction for lung cancer by annual CT monitoring of our cohort of heavy smokers. These data were confirmed by a meta-analysis conducted by two senior epidemiologists of Memorial Sloan-Kettering Cancer Center, Peter Bach and Colin Begg, of the three concurrent single arms studies: IEO/INT, Mayo Clinic and Lee Moffitt Cancer Center trials. Such a meta-analysis demonstrated that annual CT increases by 3-4 fold the number of detected lung cancers but does not reduce the incidence of advanced and potentially lethal disease. These data strengthen the validity of the plan to continue clinical research on early lung cancer detection by a randomized controlled study in 2005. In fact there is an overwhelming consensus in the International scientific community that uncontrolled observational studies will not be able to provide further knowledge in this field. Only large randomised controlled clinical trial will hopefully provide convincing evidence on the magnitude of the benefit achievable by early detection programmes with spiral CT in heavy smokers.

In 2005, with the support of AIRC (Italian Association for Cancer Research) and the Ministry of Health, the investigators launched the new INT (National Institute of Tumors) randomized controlled study, named Multicentric Italian Lung Detection trial (MILD), combining smoking cessation with early diagnosis and biologic assessment of individual risk of lung cancer.

The initial goal at INT was to recruit 10,000 subjects from 10 different Italian centres by the end of 2008. This task has proven unfeasible for a number of reasons, including the lack of available institutions with sufficient experience and logistic facilities to run large long-term trials, and the strong campaign in the media in favor of early detection programmes, which has made the randomisation in a control arm unacceptable for a large proportion of volunteers. As a matter of fact, despite the formal collaboration of Istituto Superiore di Sanità (ISS Rome), signed in February 2006, and the acceptance of the protocol by the Ethics Committee of many Italian centres, only two other Institutions have started active accrual by July 2007.

As a consequence, the INT has decided to increase its target for the MILD trial to 4000 subjects, with the aim of joining the other two Italian randomized studies (DANTE and ITALUNG) to provide a strong Italian meta-analysis of at least 10,000 subjects, that might in the future be compared with the two large on-going randomised trials, the NCI trial in the United States and the Nelson trial in northern Europe (Netherlands, Belgium, Denmark), providing concurrent data on the southern European population.

MILD study recruits subjects randomized in 2 groups: a control group undergoes to a program of primary prevention with pulmonary function test evaluation and a group to periodic spiral CT associated with primary prevention and pulmonary function test evaluation. The last one is randomized in two arms: yearly low-dose CT vs CT every 2 years. MILD trial comprehensive design combines for the first time primary prevention with early detection and molecular risk profiling representing an innovative research against lung cancer.