After the publication in 1999 of the results of the ELCAP study one arm studies have shown the good performance of low dose CT scan screening for the detection of early lung cancer cases and high survival rates, randomized trials were launched to answer the question of the efficacy and cost-effectiveness of lung cancer screening in high risk subjects. One arm studies are not sufficient mainly because the overdiagnosis related to the possible detection of indolent lung cancers detected at screening by chest Rx in older randomized trials. For this reason lung cancer screening randomized trials (RT) have been implemented in USA and in Europe. Recently the NLST US study (about 50.000 subjects) was early stopped after 8 years of follow up because an interim analysis showed a 20% mortality reduction for the intervention arm. In Europe an international collaboration between RTs was set up in Liverpool in 2005 with the aim of pooling data from national studies. Several RTs are currently under way in Europe (EULCT): NELSON (NL, B), DLCST (DN), ITALUNG (I), DANTE (I), LUSI (D), MILD (I), for a total of more than 25.000 subjects enrolled; one trial is in preparation in UK (UKLS). The largest is the Dutch-Belgian trial (NELSON), the Danish (4.000), the ITALUNG (3000). These studies compare the intervention arm, receiving annual CT Scan for different number of rounds, to the usual care, not considering early diagnosis of lung cancer appropriate according with guidelines, and DANTE (4000) and the MILD study (comparing 1 versus 2 years interval between tests). A study is still enrolling subjects in Germany and a pilot study is starting in UK (UKLS− one shot design).

In a recent meeting held in Pisa (Italy) the EULCT investigators stated the studies should continue and evaluate the full effect of screening with low-dose CT scan compared with non-screening (usual care), in terms of both mortality reduction and harmful side effects. The EULCT investigators decided to do a feasibility evaluation of a combined interim analysis of the European randomized trials during 2012, while the trials will continue until the planned end. In fact, there are many questions still open before lung cancer screening with Low Dose CT scan can be recommended. Additionally, a recommendation in favor of screening for lung cancer should not rely on one single study. Despite randomization, population and study design play a role in epidemiological studies and affect the effect estimation. The ITALUNG study has now completed the whole screening cycle (baseline +3 screening rounds) and results for the screened subjects are here reported.

Materials and Methods: The study was carried out in three screening centers in Florence, Pisa and Pistoia of the Tuscany Region (Italy) and supported by the Regional Health Public Authority. and approved by the Local Ethic Committee of each participating institution. The Institute for Cancer Prevention and Research (ISPO) of Florence was the coordinating centre. Subjects received a letter and consented to randomized. Eligible were subjects aged between 55 and 69 years at the time of enrolment with a smoking history of at least 20 packs–year since the last 10 years. Subjects who attended the baseline test (n=1406) were re-invited at subsequent 3 rounds.

CT Scan had double reading by radiologists and positive screening test for nodule at baseline was considered a solid or part-solid non calcific nodule (NCN) with mean diameter more than 5 mm. at baseline test (3 mm. at repeated screening test. The protocol was basically similar to that of the International-ELCAP Study. FDG-PET, FNA or 3-months follow up were suggested for assessment. On case-by-case basis FBS was performed. All the surgically removed lesions were evaluated according to the World Health Organization Criteria. A parallel collection of sputum and blood was obtained by the screening participants.

Results: Results of baseline and all the three annual repeat CT scans will be reported.

From 2004 to 2006, 1593 subjects were randomized as controls and 1613 enrolled in the intervention arm. 1406 (87.2%) accepted to be screened and underwent baseline CT scan. Out of the screened at baseline (1,406), 97.8%, 96.7% and 97.2% were screened at the first, second and third subsequent rounds, respectively for a total compliance of 78.3%. 426 participants (30.3%) had a NCN detected at baseline CT, and 619 (15.8%). 40 primary lung cancers were detected in 38 participants (2.7%), 21 at the baseline screening (DR: 1.4%), and 17 (DR:0.4%)at the subsequent screening rounds.

The adenocarcinomas were the 48% at baseline, and 71% at repeated tests. 6 subjects were diagnosed after the follow-up CT Scan (29%) at baseline; 12 lung cancers were diagnosed after a follow-up at repeated tests (71%).6.4% of the subjects with NCNs, had a FDG-PET; with a positive result in 29/90 subjects (32.2%) and the 22.2% of lung cancers. In 34 subjects (2.4%) work-up was carried with 38 CT guided FNA; 24 FNA (24/34= 70.5%) were positive for lung cancer and 2 for metastasis. Pneumothorax occurred in 11/38 (30%) of the subjects, but only 2 cases needed thoracic drainage. The combination of FNA and PET positive result lead to a 88.2% of subjects preoperatively classified as lung cancer. Surgery was the main therapy for 28/32 NSCLC and for 3/3 typical carcinoid and 1/1 for mixed cancer NSCLC + SCLC; 6 cancer were treated with adjuvant therapy.

Discussion: Results of the European trials will in the next years contribute to the valuation of the efficacy and performance of CT Scan Lung cancer screening. Several questions to be answered-as stated in the Pisa consensus statement-in the next future before considering low-dose CT Scan for lung cancer screening:

  1. What is the adequate target population for screening and how to assess individual risk for lung cancer, with the aim of selecting a raised risk population?

  2. How the use of early biomarkers would support the assessment of the individual lung cancer risk.

  3. What is the optimal management of CT screen detected suspicious nodules?

  4. What is the optimal protocol for screening (age range, interval, recall)

  5. What is the effect size of over-diagnosis

  6. What is the cost-effectiveness of CT screening

  7. What CT screening will add to an anti-smoking policy.

It is essential to consider together the development of primary prevention-especially smoking cessation and other anti-smoking policies - and screening implementation also modeling the expectations from different policies.

In conclusion, the European trials should continue until the scheduled end, and demonstration projects, multi-centric and possibly coordinated at European level, should be implemented aiming to evaluate essential aspects of the CT Scan lung cancer screening process. At individual level, spontaneous access to CT Scan lung cancer screening should be discouraged at the moment, considering the available evidence still insufficient to suggest its use outside a controlled research setting. In any way, exposure of subjects to CT scanning for lung cancer in spontaneous screening requires a careful information about the limited information on a potential benefit on the one side and known harmful side effects of screening on the other side, and a written informed consent when subjects volunteer to screening.

Citation Information: Cancer Prev Res 2011;4(10 Suppl):SS03-03.