Abstract CN06-03: Disease dynamics: Cancers evolving fast and slow or not at all

Abstract

Abstract Linear models of cancer progression have dominated medical thought for more than a century (Crosswell, Ransohoff, and Kramer Semin Oncol 2010; 37:202). The concept that cancer develops by a gradual, linear process that occurs over prolonged periods has inspired hope for improved cancer outcomes and decreased mortality by early detection. However, screening tests for many cancers have resulted in “overdiagnosis” of indolent conditions that cause neither death nor symptoms over a lifetime (e.g., some prostate cancers). There is strong evidence that some screening tests reduce mortality in randomized trials (e.g., fecal occult blood testing for colon cancer). However, other screening tests consistently fail to detect cancers at an early, curable stage. For example endoscopic screening of patients with reflux to detect patients with Barrett's esophagus (BE) for early detection fails to detect 90% or more of esophageal adenocarcinomas (EA), which then present as advanced, lethal malignancies (“underdiagnosis”). These outcomes are usually represented conceptually by linear diagrams in which the x axis is time and the y axis is some representation of tumor progression such as size. These observations and lack of critical data on the temporal and spatial course of progression have led to the question “Will everyone develop covert cancer?” (Greaves, Nat Rev Cancer; 2014; 14:209) and a call to study “disease dynamics” (Welch and Black, JNCI 2010: 102:613; Li et al PLoS Comp Biol 2011 Feb;7(2):e1001087). Genomic advances allow preliminary estimation of disease progression tempo even though the cancers are typically assessed only at a single point in space and time. These studies have provided evidence that some cancers appear to develop by a series of rapid (“punctuated”) or even sudden (“catastrophic”) genomic events whereas others evolve grandually (Baca et al Cell 2013; 153: 666-677). However, the representation of time on the x axis is typically vague providing no clinically relevant information. A recent manuscript reported results from a case-cohort study using 1M SNP arrays to evaluate genomic alterations in at two cm intervals in biopsies from BE in 248 patients, including 79 who progressed to EA and 169 who did not (Li, Galipeau, Paulson et al Cancer Prev Res 2014;7:114) in which biopsies were obtained at 2 cm intervals in the BE segment at the baseline and penultimate endoscopies (endoscopy before esophageal adenocarcinoma (EA) in “progressors” and next to last endoscopy in non-progressors). This study found that BE that progressed to EA was characterized by punctuated chromosome instability with gains or losses of whole chromosomes or large regions of chromosomes that became significant 48 months before EA diagnosis followed by catastrophic genome doublings in the 24 months before cancer. In contrast, BE that did not progress to EA was largely characterized by somatic genomic “stasis” at the level of 1M SNP arrays for prolonged periods up to more than two decades. These results suggest a four year “window of opportunity” for early detection although future research may expand this window. These findings are supported by recent results based on genomic studies of advanced cancers that suggest that chromosome instability followed by genome doublings may be common across many cancer types in addition to esophageal, including breast, lung, colon, and ovary, and that nearly 40% of all human cancers have undergone a genome doubling event (Carter et al Nat Biotechnology 2012; 30: 413; Zach et al Nat Genetic 2013; 45: 1134). Well-designed studies of cancers evolving in space and time are needed to determine the timing of genomic alterations, including chromosome instability and genome doublings, relative to the onset and progression of cancers in addition to EA. Citation Format: Brian J. Reid. Disease dynamics: Cancers evolving fast and slow or not at all. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr CN06-03.