Overall and non-lung cancer incidence and mortality in the National Lung Screening Trial: Opportunities for multi-cancer early detection

Background: Currently recommended cancer screening programs address only part of the overall population cancer burden. Even populations deemed high-risk for certain individual cancers experience a considerable potential burden of other cancers. However, few published cancer screening trials report the incidence of untargeted cancers. Methods: The National Lung Screening Trial (NLST), initiated in 2002–2004, was a randomized controlled trial of lung cancer screening in adults with ≥30 pack-years of smoking. Active follow-up for incident invasive cancers continued through 2009. Results: Among 53,229 NLST subjects (median follow-up 6.5 years after randomization), the incidence of lung cancer was 615 per 100,000 person-years (32% of 6142 overall first primary incident invasive cancers), and that of non-lung cancer was 1327 per 100,000 (68%). Non-lung cancer incidence exceeded that for lung cancer in all 5-year age categories and all quintiles of smoking pack-years. Besides lung cancer, the most common cancers were prostate, breast, colon/rectum, bladder, and head/neck; 23% were smoking-related cancers, and 54% were cancer types lacking recommended population-based screening modalities (32% excluding prostate). Non-lung cancer comprised 48% of 1793 cancer deaths. Conclusions: In the NLST, only 32% of first primary cancer incidence after study entry was lung, compared with 68% non-lung. Even in a population at high risk for lung cancer, a single-cancer screening test misses most cancers. Thus, in combination with existing single-cancer screening modalities, multi-cancer screening tests—which address many of the incident non-lung cancers in this trial—have potential to address a currently inaccessible portion of cancer morbidity and mortality.