Long Article

Benefits of Extending Lung Cancer Screening Increases at 10 Years

Results from the Multicentric Italian Lung Detection (MILD) study, the only randomized lung cancer (LC) screening trial designed to evaluate extended intervention, delivered further corroborating evidence that low-dose computed tomography (LDCT) LC assessment for high-risk populations improves survival.1 The study confirmed a 10-year improved survival benefit of 39 percent by extending LC screenings that may increase the chance for early detection.1

With lung cancer ranking as the second most common cancer in the United States as well as the leading cause of cancer-related deaths, these results are positive news given treatment may be more successful when the disease is found at an early stage and before it has spread.1,2,3 Lung cancer, estimated at $13 billion in annual healthcare costs, is usually detected and diagnosed in late stage after metastasis has occurred and it is difficult to treat.2,3 Early detection and treatment that may be achievable with prolonged screening could give patients a better quality of life and the chance to live longer.2,3

Results from MILD complement those of an earlier U.S. trial conducted at 33 sites, the National Lung Screening Trial (NLST), which reported an increased survival benefit of 20 percent based on a protocol of three annual LDCT screenings compared to a chest radiography control group.1

Enrolling high-risk patient groups

Launched in 2005, MILD was a multicenter prospective randomized controlled lung cancer screening study that enrolled more than 4,000 current and former smokers who had quit within the past 10 years and no history of cancer during the previous five years.1 Ranging in age from 49 to 75, each person was randomly assigned to an intervention arm and received LDCT screening or the control arm with no intervention.1


As a secondary aim, MILD compared the effectiveness of two different LDCT screening intervals based on the hypothesis that increased screening intervals could help reduce costs and radiation dose without interfering in the benefits of screening and early diagnosis.1 To evaluate this theory, patients in the LDCT screening arm were further randomized into annual (every 12 months) or biennial (every 24 months) LDCT groups.1

For both the MILD and Nederlands-Leuvens Longkanker Screenings ONderzoek (NELSON) trial, longer screening intervals were used but with different approaches.1 MILD randomized patients to two screening interval groups of 12 or 24 months and tested them both throughout the duration of the study period.1 By contrast, NELSON evaluated three consecutive rounds of scanning with varying longitudinal intervals for the same participant.1

Results of the final round of scans from the NELSON trial found more than twice the occurrence of stage III/IV LC and five times the occurrence of interval cancer when compared with the first year.1 While the MILD study found a similar proportion of stage I disease, resections, and interval cancers between the 12 and 24 month LDCT screening groups, it was accompanied by lower costs and reductions in exposure to radiation.1

Screening for MILD participants continued for more than six years with more than 93 percent reaching nine years of follow-up and 71 percent of patients accumulating 10 years of follow-up.1

Compensating for early detection risks

Risks of early detection LC screening that have been documented by previous trials include overdiagnosis and overtreatment of benign tumors with unnecessary surgery.1 Reported occurrence of nonessential treatment in screening trials has ranged from more than 18 percent (NLST) to over 67 percent (Danish Lung Cancer Screening Trial - DLCST).1

To address this risk, MILD researchers developed a customized protocol that involved active surveillance of subsolid lesions for slow growing nodules.1 Additionally, selective use of PET scans improved differential diagnosis and resulted in a less than five percent resection rate for benign tissue.1 This resection rate was a significant improvement compared to the more than 24 percent rate reported by the NLST trial or the 15 percent threshold recommendation from the National Comprehensive Cancer Network (NCCN).1 A comparable resection rate of less than six percent was also reported in the MILD control arm where patients received no screening intervention.1

Extended screening detects treatable disease

Early stage LC is considered more treatable because patients are more likely to qualify for surgical intervention.1 During the trial, investigators detected 158 cases of lung cancer with 98 (or 62 percent) of those participants in the screening group.1 Of those 98 patients, 49 (or 50 percent) were diagnosed with stage one disease and 64 (or more than 65 percent) qualified for and underwent resection.1

For the 60 patients (or 38 percent) in the control arm receiving no screening intervention, they were less likely to be diagnosed with stage one (22 percent) lung cancer and only 13 patients (22 percent) qualified for resection.1 Ultimately, lung cancer accounted for 29 percent of deaths in the LDCT group compared to 38 percent in the control arm.1

Benefits increase over time

In addition to the reported lung cancer survival benefit 39 percent at 10 years with LDCT screening, a 20 percent reduction in overall mortality for the intervention group was also confirmed.1 Risk for death attributable to LC at 10 years was calculated at 1.7 percent in the intervention group compared to 2.5 percent in the control arm.1 In comparison, the 10-year risk for overall mortality was 5.8 percent with screening compared to 6.5 without.1

Landmark analysis of MILD outcome data revealed that the benefit of LDCT screening could extend beyond five years, with statistically significant reductions of 58 percent for LC mortality (0.7 vs 1.5 percent risk) and 32 percent for overall mortality (3.4 vs 4.5 percent risk).1 Additionally, a sensitivity analysis that encompassed a more homogeneous group of nearly 3,500 study participants found a 49 percent reduction in lung cancer mortality for LDCT screening beyond five years.1

Comparing previous trial outcomes

According to MILD investigators, patient enrollment criteria for other randomized controlled trial (RCT) were not comparable and most European trials enrolled younger populations with lower risk.1 Additionally, most RCTs performed annual LDCTs for four years or less so that evaluating the impact of screening duration and intensity on mortality was not feasible.1

When comparing MILD results for extended LDCT screening intervention to the results for NLST, which evaluated LDCT benefits for three annual scans compared to chest radiography beginning at baseline, there was a 95 percent increase (39 vs. 20 percent) in the survival benefit for deaths attributable to LC when screening was prolonged.1 Additionally, the different risk profile implemented by the NLST trial would have only allowed 51 percent of MILD participants to qualify and helps clarify the causes for the difference observed in the lung cancer death rate of control groups in the NLST (309 vs. 247 per 100,000 person-years) compared to MILD. Further, the 10-year lung cancer mortality rate reduction was greater for the MILD screening arm (173 vs. 247 per 100,000 person-years) when compared to the NLST screening group.1

Researchers reported that that one case of lung cancer was identified for every 154 LDCTs and 1.4 PET scans performed.1 Other analysis revealed that 167 screened patients, 733 LDCTs, and 4.4 PET scans used for differential diagnosis were needed to prevent one death attributable to lung cancer.1


  1. Prolonged lung cancer screening reduced 10-year mortality in the MILD trial: new confirmation of lung cancer screening efficacy. Annals of Oncology https://doi.org/10.1093/annonc/mdz117  Accessed 5/27/2019
  2. Can Non-Small Cell Lung Cancer Be Found Early? American Cancer Society https://www.cancer.org/cancer/non-small-cell-lung-cancer/detection-diagnosis-staging/detection.html Accessed 5/27/2019
  3. Lung Cancer Fact Sheet. American Lung Association https://www.lung.org/lung-health-and-diseases/lung-disease-lookup/lung-cancer/resource-library/lung-cancer-fact-sheet.html Accessed 5/27/2019