Colorectal cancer (CRC) incidence has been increasing at a concerning rate in the United States among individuals age 55 and younger, and the reasons why remain unclear. A General Session held January 20 during the 2018 Gastrointestinal Cancers Symposium provided a comprehensive overview of early-onset CRC, including results of a study focused on the age distribution of tumor gene expression in patients with stage II/III colon cancer.
Joseph J.Y. Sung, MD, PhD, MBBS, of the Chinese University of Hong Kong, in China, began with data illustrating that although the overall incidence of CRC in the United States demonstrates a steady decline since the mid-1970s, this trend has been disrupted in select subgroups. Beginning in the mid-1980s, the incidence of colon cancer has increased by 1.0% to 2.4% per year in adults age 20 to 39 years and by 0.5% to 1.3% per year in adults age 40 to 54 years.1 The changes in the rates of rectal cancer incidence are even more pronounced—increasing 3.2% per year since 1974 in adults age 20 to 29 years. Notably, the proportion of rectal cancer cases that were diagnosed in adults younger than age 55 doubled from 14.6% between 1989 and 1990 to 29.2% between 2012 and 2013.
Dr. Sung underscored data illustrating that U.S. individuals born around 1990 face a doubled risk of colon cancer (incidence rate ratio [IRR] 2.40, 95% CI [1.11, 5.19]) and a quadrupled risk of rectal cancer (IRR 4.32, 95% CI [2.19, 8.51]), as compared with adults born around 1950.1
A panel discussion takes place during General Session 7.
All told, collective epidemiologic data suggest that the increases in early-onset CRC appear to be more prominent in men than women and in rectal cancer rather than colon cancer. Moreover, this is a phenomenon occurring around the world including in Europe and Asia, not just in the United States.
The incidence of CRC begins to become apparent in individuals starting around age 50 and increases with age, which is why most screening programs recommend that screening be started at age 50. However, based on the data presented, particularly the alarming increases in rectal cancer in individuals younger than age 55, Dr. Sung put forth the question of whether it is time to consider screening for CRC in younger age groups.
Screening, Prevention, and Health Disparities
Fay Kastrinos, MD, MPH, of Columbia University Medical Center, addressed the issue of earlier screening during her presentation. Since the launch of population-wide screening, the incidence of CRC has declined in the United States, with the greatest reductions in the population age 50 and older. Currently, however, approximately one of every 10 individuals who develops CRC is younger than age 50. Many have questioned whether the United States should initiate CRC screening before age 50 to catch CRC early in this small proportion of individuals.
Numerous modeling analyses evaluating the optimal starting age and screening intervals for various screening tests have been performed. However, the discordant findings across models, the lack of empiric evidence in younger populations, and the modest increase in life-years gained with earlier screening do not support the benefit of earlier screening according to the U.S. Preventive Services Task Force. Of course, these recommendations could change with the introduction of better-accepted, more cost-effective screening tools.
An obvious question is whether the increasing incidence of early-onset CRC reflects a true increase in disease burden or increased disease awareness and early diagnosis. Although colonoscopy rates tend to parallel increases in CRC incidence up until 2009 when the U.S. recession hit, Dr. Kastrinos noted that the decline in colonoscopy rates after 2009, in conjunction with the increasing incidence of early-onset CRC, possibly suggest that additional factors—beyond detection and diagnosis—contribute to earlier development of disease. These potential factors are currently undefined. Numerous risk factors were proposed—for example, obesity, a sedentary lifestyle, increased consumption of processed meat, and microbiome differences—but all require further study.
It is currently unclear whether distinct mechanisms might drive CRC onset and progression in racial subgroups, according to Dr. Kastrinos. A higher incidence of CRC has been found among young black and Hispanic populations. Moreover, emerging data suggest that differences among these racial groups in tumor site, stage at diagnosis, and histology may reflect racial variations in CRC risk factors, such as biology and differential exposure to modifiable risk factors, rather than disparities in diagnosis and treatment. Data on the effectiveness of different screening strategies for diverse populations are not available, but some of these subgroups may benefit from screening before age 50.
Genetic analyses point to a high frequency of germline mutations (16%), as well as a wide spectrum of mutations among individuals who develop early-onset CRC. Dr. Kastrinos suggested that universal multigene germline testing for all individuals with early-onset CRC would be worthwhile for guiding genetic counseling and familial testing.
Clinical Management and Survivorship Issues
Irit Ben-Aharon, MD, PhD, of the Davidoff Cancer Center, explored three domains relevant to patients with early-onset CRC: treatment-induced sequelae, psychosocial needs, and the relevance of unique cancer biology.
Treatment-related toxicities weigh heavily on many young cancer survivors, and fertility represents a prime concern, particularly for women. Dr. Ben-Aharon shed light on this issue for patients with CRC. A small retrospective study found that FOLFOX-induced amenorrhea during chemotherapy tends to increase with age among premenopausal women. Limited early data suggest that oxaliplatin-based protocols may be less gonadotoxic than other chemotherapy regimens. Although gonadotropin-releasing hormone analogs offer protective effects in the setting of certain chemotherapies, Dr. Ben-Aharon noted that there is a lack of evidence supporting these agents in combination with current CRC regimens.
Other work suggested that survivors treated with abdominal radiotherapy faced a 3-fold higher risk of delivering offspring of low birthweight compared with survivors treated without radiotherapy, although there is a lack of evidence regarding the uterine toxicity caused by new radiotherapy techniques.
Dr. Howard S. Hochster speaks during General Session 7.
Given the scant literature on the effects of various anticancer treatments in younger individuals with gastrointestinal malignancies, Dr. Ben-Aharon indicated that more study is needed to better counsel patients regarding fertility preservation, quality of life, and survivorship issues. Toward this end, the European Organisation of Research and Treatment of Cancer has established the Young-Onset Colorectal Cancer Task Force to develop prospective studies that focus on these very issues.
Dr. Ben-Aharon also touched briefly on the issue of early-onset CRC biology, noting that early-onset cancer is a hallmark of inherited cancer predisposition. Yet, as Dr. Kastrinos showed, only about 10% to 15% of early-onset CRCs are hereditary.
“Elucidating the potential unique biology is essential for potentially superior treatment tailoring in the future,” Dr. Ben-Aharon said. She believes the time is right to begin leveraging state-of-the-art genomics technology to analyze and interpret the molecular profiles of early-onset CRCs at the DNA, RNA, protein, and epigenetic levels and has started working to do so. A small exploratory study carried out by Dr. Ben-Aharon and her colleagues suggest that somatic gene mutations and methylation patterns vary considerably between patients with early-onset CRC versus average-onset CRC. Most notably, they found that many genes were hypomethylated in early-onset CRC and that these genes demonstrated increasing methylation with advancing age beyond naturally occurring variation in matching normal tissue.
Gene Expression Patterns in Colon Cancer Based on Patient Age
Howard S. Hochster, MD, of the Yale School of Medicine, and colleagues have started to address this very issue. To determine whether early-onset CRC is biologically distinct from that arising in older individuals, the investigators evaluated whether age-specific differences could be identified using a well-validated multigene expression profile, the 12-gene Oncotype DX Colon Recurrence Score test, which is used to predict the risk of recurrence in patients with stage II/III colon cancer (Abstract 552).
After assessing more than 22,000 samples from patients with stage II/III colon cancer, the findings revealed no differences in gene expression patterns across all age groups (i.e., age < 40, 40 to 54, 55 to 64, and ≥ 65; age < 55 and ≥ 55).
“Our findings suggest that colon cancer in younger [patients] with resected stage II and [disease] is not biologically different from that in the older patients, as measured by this test,” Dr. Hochster concluded.
The results showed that the majority (70% to 73%) of patients across all age groups were categorized as having low-risk disease, including the younger patients under age 55. The prevalence of high-risk disease was identical across all age groups at 8%.
Dr. Hochster remarked that the findings underscore the importance of identifying the risk of recurrence in younger patients, many of whom have low-risk disease. Given their age, many clinicians choose to manage these individuals aggressively by administering adjuvant chemotherapy. However, the recurrence-risk data suggest this might represent overtreatment for many individuals.
Data for the 22,052 patients with stage II and stage III A/B colon cancer comprising the analysis were obtained from the Genomic Health, Inc. clinical laboratory. Nearly one-quarter (23%) of the patients in the cohort were younger than age 55, and 93% have stage II disease.
– Kara Nyberg, PhD