Dr. Brian J. Reid

CLONAL EVOLUTION, CANCER PREVENTION AND CANCER RISK PREDICTION IN BARRETT'S ESOPHAGUS

"Acquired genetic lability permits stepwise selection of variant sublines and underlies tumor progression."
Peter Nowell, Science 1976; 194:23

http://seattle.barrettsresearch.org/images/translation.gif

As an undergraduate chemistry major, I became interested in the potential for genetics to address complex biological problems. I had the great fortune to enter Lee Hartwell's lab for an undergraduate research project, and in early 1969, we discovered yeast cell cycle mutants. After a brief stint at MIT, where I had been accepted as a graduate student, I returned to become part of the team that proposed the first genetic model of the eukaryotic cell cycle, including a G1/S regulatory element called START. This experience was one of the richest in my life and taught me numerous lessons, including the value of team science, the importance of a model system and the wisdom of choosing an important problem that was worth devoting my career to studying. At the time, two other members of the Department of Genetics, Stan Gartler and Phil Fialkow, taught me emerging concepts of neoplastic clonal evolution, later summarized eloquently by Dr. Peter Nowell. This appealed to me. It was easy to hypothesize that genes affecting a START like function could lead to cancer, neoplastic evolution was a significant problem and addressing it well could help patients. I entered medical school, where I realized that the recent innovation of fiberoptic endoscopy would permit study of neoplasms at their earliest stages of progression. I eventually selected Barrett's esophagus as a model of human intraepithelial neoplasia.

At the time, Barrett's esophagus and esophageal adenocarcinoma were thought to be extremely rare conditions. However, the incidence of esophageal adenocarcinoma is now known to be increasing at an alarming rate in many western countries, rising nearly 600% in the US over the three decades from 1972 - 2002. Esophageal adenocarcinoma is highly lethal with five year survival of less than 15%, and treatment is associated with significant morbidity and mortality depending on the experience of the institution.

Barrett's esophagus is the only known precursor to esophageal adenocarcinoma with an estimated annual incidence of 0.5% to 1.0%, but Barrett's esophagus is also an indolent disease that will not progress to cancer during the lifetimes of most patients. Thus, these patients are between a rock and a hard spot; if an advanced esophageal adenocarcinoma develops, they face greater than 85% mortality, but while most will never get cancer, they are yet exposed to repeated, invasive and expensive procedures because the benign subset cannot be identified with precision. Two advances are greatly needed to achieve the NIH goals of "personalized medicine": improved cancer risk prediction and low-cost, well tolerated interventions to prevent cancer.

A patient's course will be modulated by host and environmental risk and protective factors as well as generation of variants, natural selection and neoplastic evolution of clones within the Barrett's segment. In the 1980s, the Seattle Barrett's Esophagus Team developed the endoscopic biopsy protocols and the dysplasia classification system that are the current standard of care for cancer risk stratification in Barrett's esophagus. We evaluated observer agreement in a blinded study, and we and others subsequently performed prospective studies of dysplasia to determine how well it predicted progression to cancer in Barrett's esophagus. The results indicated potential for improvement because dysplasia had a number of limitations, including (1) being subjective with substantial interobserver disagreement in diagnosis; (2) having large biological heterogeneity with regard to cancer outcome; (3) its focal nature that required large numbers of biopsies for detection, and (4) absence of low-cost, well tolerated interventions linked to risk. During the late 1980s and early 1990s, we established the scientific base for studying clonal evolution longitudinally that included low-density, genome-wide allelotype studies that identified a limited number of regions of selected LOH, including 9p LOH and 17p LOH; genetic dependency studies among the selected abnormalities to determine the order in which they developed in neoplastic progression; a clinical team presently led by Dr. Patricia Blount responsible for research and safety for the Seattle Barrett's Esophagus Cohort of more than 600 patients, which has become one of the best characterized in the world; biorepositories; and an epidemiology study of the cohort, led by Dr. Thomas Vaughan. We have two laboratories studying neoplastic evolution, including the Reid Lab that studies clonal evolution and the Rabinovitch Lab that studies genomic instability. We also have a computational team that includes biostatisticians, evolutionary biologists and computer scientists lead by Drs. Carlo Maley (Wistar Institute) and Xiaohong Li (Fred Hutchinson Cancer Research Center). In summary, at present we have assembled a multidisciplinary, inter-institutional team that includes three projects (epidemiology, genetic instability and clonal evolution) and three cores (leadership, clinical research, and biostatistics and evolutionary analyses) to study predictors of progression in Barrett's esophagus. We have 24 years of experience in early translational research studies as defined by the NCI Translational Research Working Group (http://www.cancer.gov/trwg/TRWG-definition-and-TR-continuum).

Recent research advances

DNA chromosome instability cancer risk prediction panel. Based on discovery research in the first half of the 1990s, we hypothesized that abnormalities involving two tumor suppressor genes (CDKN2A {p16}, TP53) and DNA content (tetraploidy, aneuploidy) would improve cancer risk prediction. Currently, we are completing analysis of a prospective cohort study of patients with Barrett's esophagus followed for up to 11 years. The results from this study indicate that 9p (p16) LOH, 17p (p53) LOH, tetraploidy and aneuploidy, detect high- and low-risk patients with much greater precision than has been previously possible (Galipeau et al PLoS Medicine 2007; 4: 342-54). A panel of these abnormalities was the best predictor of progression to esophageal adenocarcinoma (RR = 38.7; 95% CI = 10.8 - 138.5; p<0.001). Patients who had 9p LOH, 17p LOH and a DNA content abnormality (tetraploidy or aneuploidy) had a 79.1% five year cumulative incidence of cancer. In contrast, patients with none of these abnormalities at baseline had a 0% cumulative incidence of cancer to nearly 8 years (95 months). Other abnormalities of these tumor suppressor genes, including promoter hyper-methylation or mutation, did not provide independent predictive information beyond that provided by LOH and DNA content. This panel makes advances relative to all the limitations of the current classification system because the panel (1) is based on objective statistical calls of abnormalities rather than subjective interpretations, (2) improves detection of both high and low risk populations, (3) requires only 1/4 to 1/8 or fewer biopsies as dysplasia, and (4) provides stratification of patients at greatest risk of cancer and links a low-cost, well tolerated intervention to somatic genetic alterations, as described below.

http://seattle.barrettsresearch.org/images/studytimeline.jpg

Clonal evolution may be modulated by host and environmental factors. We also found that host and environmental factors appear to modulate clonal evolution during neoplastic progression in Barrett's esophagus, identifying a subset of Barrett's patients who are at increased risk for developing chromosomal instability (Chao et al CEBP 2006; 15: 1935-40) and finding evidence that a low-cost, well tolerated intervention, regular use of aspirin or other NSAIDs, might prevent esophageal adenocarcinoma as well as progression to tetraploidy and aneuploidy (Vaughan et al Lancet Oncol 2005; 6: 945-52). Further, this protective association was especially strong in high-risk patients who had already developed multiple abnormalities at baseline. (Galipeau et al PLoS Medicine 2007; 4: 342-54). For example, the strongest protective association was seen in patients with multiple genetic abnormalities, with NSAID nonusers observed having a 10-year esophageal adenocarcinoma risk of 79% compared to 30% for NSAID users (p<0.001).

Selectively advantageous mutations and hitchhikers: p16 lesions are selected in Barrett's esophagus. Neoplastic evolution is characterized by epigenetic and genetic instability and waves of clonal expansions carrying genetic and epigenetic abnormalities to fixation (100% of the cell population). However, an evolutionarily neutral abnormality can expand to fixation if it spreads as a hitchhiker ("passenger") on a selective sweep. We reported that p16 LOH, methylation and mutation had strong and independent effects, driving clonal expansion early in progression (p<0.001) (Maley et al Cancer Res 2004; 64: 3414-27). Second lesions in p16 and TP53 (LOH, mutation) are associated with later selective sweeps (p<0.001). All other abnormalities, including microsatellite shifts and LOH involving other loci, could be explained as hitchhikers on p16 mediated clonal expansions. To our knowledge, this was the first large-scale study to distinguish between selected and hitchhiker abnormalities in any neoplasm, and it illustrates the utility of the genetic dependency analyses that we have previously combined with robust biostatistical approaches to select candidate genes and genome regions for our research priorities of cancer risk prediction, prevention and early detection.

The combination of clonal expansion and genetic instability predicts progression to esophageal adenocarcinoma. There has been a debate over the relative importance of clonal expansion and genetic instability during progression to cancer. To address this question, we conducted a prospective investigation of 267 patients with BE in whom biopsies were characterized for p16 (LOH, mutation, methylation), TP53 (LOH, mutation), DNA content tetraploidy and aneuploidy (Maley et al Cancer Res 2004; 64: 7629-33). We reported that the risk of EA increased exponentially with the size of TP53, tetraploid and aneuploid clones (RR=1.33n for an n cm clone, 95% CI: 1.15-1.55). Clone size was a better predictor than the mere presence or absence of the clones. The size of a clone with p16 abnormalities was not an independent predictor of progression when TP53 status was considered, and controlling for BE segment length had no significant effect on the results. We found that the most important measure of neoplastic cell population size was not the total number of BE cells, but the number of genetically unstable BE cells. Thus, the combination of genetic instability with clonal expansion seems to drive BE progression, not one or the other alone. The results imply that interventions that limit expansion of chromosomally unstable clones may reduce risk of EA. NSAIDs are a prime candidate for such an intervention.

Clonal genetic diversity predicts progression to esophageal adenocarcinoma. There have been major advances in our understanding of molecular mechanisms of genetic instability, inactivation of tumor suppressor genes and activation of oncogenes. However, few studies have investigated clonal evolutionary mechanisms that drive neoplastic progression. For example, it was not known whether homogenization of a neoplasm as a result of a clonal expansion or accumulation of clonal diversity as a measure of viable genetic variants is more predictive of progression to cancer. Mutation rates have been difficult to study in vivo. Measures of genetic diversity are indirect measures of mutation rate in that they assay the rate at which viable mutations arise and clones carrying those mutations expand to a detectable size. In this study, we reported that clonal diversity assessed by measures adapted from evolutionary biology and ecology predicts progression to EA (Maley et al, Nature Genetics 2006; 38: 468-473). The number of genetically distinct clones (RR = 1.43 per clone, 95% CI: 1.16-1.72), and the genetic divergence between clones (RR =1.62 per 10% divergence, 95% CI: 1.15-2.27) were independent predictors of progression, even when we controlled for the presence of TP53 LOH, aneuploidy and tetraploidy.

Current research

Collectively, our recent research indicates that Barrett's esophagus, which is typically classified as an intestinal metaplasia, also fulfills the definition of a neoplasm. The Barrett's neoplasm is characterized by clonality and heterogeneity even at its earliest stages suggesting that it might be viewed more as an evolving ecosystem rather than a homogeneous population of cells.

The roles of genetic and epigenetic instability in an evolving neoplastic ecosystem are poorly understood. Studies of DNA methylation in Barrett's esophagus have been limited to only two or three dozen candidate genes from other cancers, and no large scale genome-wide studies of chromosomal instability have been performed using modern array technology in patients with premalignant disease but no cancer. We are presently performing cross-sectional pilot studies of DNA methylation, LOH and copy number in preparation for a longitudinal study to identify important epigenetic and genetic determinants of progression and the rate at which diversity and selective sweeps develop in Barrett's esophagus.

http://seattle.barrettsresearch.org/images/NeoplasticProgression.gif

Public Service

I have become increasingly concerned about the persistent high cancer mortality in the United States in the face of more than three decades of research since passage of the National Cancer Act of 1971. For the last six years, I have devoted time each year to a variety of commissions, task forces, and working groups to promote research on esophageal cancer and to define and improve translational research in general. In 2001, I was a member of the NCI/NIDDK Barrett's Working Group. From 2001 to 2003, I served as co-chair of the NCI Stomach Esophagus Progress Review Group (SEPRG). From 2003 to 2006, I served on the AACR Task Force on Cancer Prevention as co-chair of the esophageal adenocarcinoma section and as a member of the writing committee. From 2005 to 2007, I was a member of the NCI Translational Research Working Group, helping develop the biomarker pathway and serving as co-chair of the Organization and Funding Committee. In 2006, I became co-chair of the esophageal cancer section for the National Commission on Digestive Diseases.

I've learned a lot from these experiences, including the difficulties facing patients with uncommon cancers, such as esophageal, the challenges involved in bringing scientific discoveries to the clinic, and the disparities in health care that can disproportionately affect minorities, such as those afflicted with esophageal squamous cell carcinoma. Although many of the recommendations we have made have not been implemented, there has been slow, fragile progress. The Barrett's Working Group recommendations resulted in an RFA that provided the first NIH grant for many young investigators. Although the SEPRG recommendation for a Stomach Esophagus Consortium was not implemented by the NCI, beginning in 2005, many researchers under the leadership of Dr. Wong-Ho Chow at the NCI aggregated into an international Barrett's esophagus, esophageal adenocarcinoma consortium (BEACON). Many recipients of Barrett's Working Group RFA grants also joined the consortium. Dr. Thomas Vaughan, a member of the Seattle Barrett's Esophagus Team became the first chair of the Steering Committee, of which I am a member.

Professional Experience:
1980-1981 Intern in Medicine, Brigham and Women's Hospital, Boston, Massachusetts
1981-1982 Assistant Resident Physician in Medicine, Brigham and Women's Hospital, Boston, Massachusetts
1982-1983 Senior Resident Physician in Medicine, Brigham and Women's Hospital, Boston, Massachusetts
1982-1983 Chief Resident Physician in Primary Care, Brigham and Women's Hospital, Boston, Massachusetts
1983-1985 Fellow in Gastroenterology, University of Washington, Seattle, Washington
1985-1986 Acting Instructor in Medicine, University of Washington, Seattle, Washington
1986-1988 Acting Assistant Professor in Medicine, University of Washington Seattle, Washington
1988-1993 Assistant Professor in Medicine, University of Washington Seattle, Washington
1993-1998 Associate Professor in Medicine, University of Washington, Seattle, Washington
1995-1998 Adjunct Associate Professor of Genetics, University of Washington Seattle, Washington
1996- Full Member, Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, Washington
1998- Professor of Medicine, University of Washington, Seattle, Washington
1998- Adjunct Professor of Genetics, University of Washington, Seattle, Washington
1998-2002 Program Head in Gastrointestinal Oncology, Fred Hutchinson Cancer Research Center
1999- Full Member, Division of Human Biology, Fred Hutchinson Cancer Research Center

Honors:
Summa Cum Laude, 1969
Phi Beta Kappa
Phi Lambda Upsilon
Alpha Omega Alpha
Seattle Academy of Internal Medicine Award, 1980
R. Robert and Sally D. Funderburg Research Scholar Award, 1993

Board Certification:
American Board of Internal Medicine, 1984
American Board of Internal Medicine,
Gastroenterology Subspecialty Board, 1990

Medical Licensure:
State of Washington #0020775

Professional Organizations:
American Gastroenterological Association
American Society for Gastrointestinal Endoscopy
American College of Physicians
Gastroenterology Research Group
American Society for Cell Biology
Northwest Endoscopy Society
American Association for Cancer Research
AGA Gastrointestinal Oncology Section
Western Association of Physicians

Public Service, Citizenship and Leadership:

2001 NCI/NIDDK Barrett's Working Group
Drs. Reid and Richard Sampliner presented the recommendations to the NCI Executive Committee
2001-2003 NCI Stomach Esophagus Progress Review Group - Co-chair
2003-2006 AACR Task Force on Cancer Prevention - Co-chair Esophageal Section, writing committee
2005-2007 NCI Translational Research Working Group - Co-chair Funding and Organization Committee
2005-Present Barrett's Esophagus Esophageal Adenocarcinoma Consortium (BEACON) - Steering Committee, Co-chair Genetics Committee
2006-2007 EDRN External Review Committee
2006-2007 National Commission Digestive Diseases - Co-chair Esophageal Cancer Section