Universal cancer screening: revolutionary, rational, and realizable

Abstract

Cancer remains the second leading cause of mortality worldwide, and overall cancer-related deaths are increasing. Despite the survival benefit from early detection, screening has to date targeted only those few organs that harbor tumors of sufficient prevalence to show cost-effectiveness at population levels, leaving most cancer types unscreened. In this perspective overview, a case is made for universal cancer screening as a logical and more inclusive approach with potentially high impact. The centrally important conceptual drivers to universal screening are biological and epidemiological. The shared biology of tumor marker release into a common distant medium, like blood, can be exploited for multi-cancer detection from a single test. And, by aggregating prevalence rates, universal screening allows all cancers (including less common ones) to be included as targets, increases screening efficiency and integration across tumor types, and potentially improves cost-effectiveness over single-organ approaches. The identification of new tumor marker classes with both broad expression across tumor types and site-prediction, remarkable advances in assay technologies, and compelling early clinical data increase the likelihood of actualizing this new paradigm. Multi-organ screening could be achieved by targeting markers within or stemming from the circulation (including blood, urine, saliva, and expired breath) or those exfoliated into common excretory pathways (including the gastrointestinal and female reproductive tracts). Rigorous clinical studies in intended use populations and collaborations between academia, industry, professional societies, and government will be required to bring this lofty vision to a population application.

Fig. 1

Current single-organ and future universal cancer screening approaches: a conceptual comparison of features

Fig. 2

Impact of cancer prevalence on screening efficiencies. a Exponential relationship between cancer prevalence and the number of patients needed to be screened to detect a single cancer (NNS). Estimated NNS is plotted for cancers at individual gastrointestinal organs (only colorectal screening is currently practiced), for combined gastrointestinal cancers (Pan-GI), and for all cancer types in aggregate (Universal). For this illustration, detection sensitivities of 100% were assumed in calculations of NNS. b Influence of cancer prevalence on positive predictive value (PPV) at various specificities. For illustrative purposes, estimated PPVs are plotted for same spectrum of single and combined cancer screening approaches as in a. For both a and b, conservative prevalence estimates obtained from the literature are used.^,,

Fig. 3

Detection and site prediction of surgically resectable cancers with a multi-analyte blood test: early results. Performance data from a prototype assay targeting various proteins and gene mutations in plasma are shown across eight common cancer types. a Sensitivity by cancer type at 99% specificity. b Accuracy of tumor localization in test-positive patients. Percentages correspond to the proportion of patients in whom tumor location was correctly classified as the most likely site (light bars) or as one of the two most likely sites (light + dark bars). Figures modified from the original publication.