Real-World Analysis Suggests Social Determinants of Health May Be Associated With Early Onset of CRC

Jessica Paulus, ScD

Despite several major United States (US) health organizations lowering the recommended starting age for colorectal cancer (CRC) screening to 45 years since 2018, the incidence of CRC in patients younger than 50 years of age is still on the rise, according to Jessica Paulus, ScD, who noted that social determinants of health could be contributing to earlier onset of CRC.1,2

Results from a real-world, retrospective analysis, which were presented at the 2025 ASCO Annual Meeting, demonstrated an overrepresentation of Black, Asian, and American Indian/Alaskan Native, and Hispanic patients in the early- vs average-onset group.2 Stage III/IV disease and obesity were also more prevalent in the early- vs average-onset group. Moreover, advanced stage at initial diagnosis was identified as the strongest predictor of overall survival (OS) outcomes among all patients.

“The current age [recommended for CRC screening], although it has been lowered, isn't enough. There was a lot of interesting commentary among panelists at ASCO about how aggressive we should be in lowering the [screening] age,” said Paulus, the senior director of Observational Research at Ontada in Boston, Massachusetts. “It is important to consider whether there's any risk-based screening that we should perform for patients under age 45 years or [incorporate into] the current population-based screening guidelines.”

In an interview with OncLive®, Paulus highlighted different factors associated with early-onset vs average-onset CRC, key findings from this analysis, and future research directions to establish solutions for early-onset CRC.

OncLive: Why is the rising incidence of early-onset CRC of particular concern?

Paulus: This work was motivated by what the oncology and public health community is recognizing as an epidemic of early-onset CRC, [which is defined as] CRC diagnosed before patients reach the age of 50. It's this curious pattern where we see stark increases in the rates of early-onset CRC but decreases in the rates of average-onset CRC. Everyone is kind of galvanized to figure out some of the reasons why this might be happening. [We are] also trying to figure out how to better support our patients who are diagnosed with early-onset CRC. In particular, there has been a lot of attention on potential social determinants of health that might be underlying increasing rates of CRC in young people, as well as [factors] that could influence survival outcomes for patients with early-onset CRC. We sought to examine this with a big data lens. [In this study], there were 14,611 patients with early-onset CRC, [who were compared with] patients with average-onset CRC, [totaling] over 100,000 patients with CRC. [It was] a statistically powerful way to try and unravel some of these questions.

What factors were evaluated in the early-onset vs average-onset patient populations?

This was a real-world database observational cohort study of patients over the age of 18 who were treated for CRC within the US Oncology Network, as well as non-network practices. The common factor [between these groups] is that [practices all utilized] the same electronic medical record, known as iKnowMed, which created a platform to generate electronic health record data that we used for research purposes. We ended up comparing social determinants of health and clinical factors at the baseline diagnosis, and then also examined survival outcomes, such as OS, in the early-onset patients [vs] the average-onset patients.

What were the key findings from this analysis?

There were 3 main findings from our work. The first is that we saw an overrepresentation of Black, Asian, American Indian, and Alaskan Native patients [in the early-onset group] as compared with the average-onset patient population. We also saw an increasing prevalence of obesity at diagnosis among the early-onset patients as compared with the average-onset patients. Perhaps most concerning for outcomes in this population is that we saw that the [patients with] early-onset [CRC] had an approximately 10% greater likelihood of advanced stage of disease at diagnosis, [specifically] stage III or stage IV disease, as compared with the average-onset patients, which is significant for these patients.

Our last analysis showed that stage at diagnosis, and stage IV disease in particular, is the strongest predictor of OS in all patients, regardless of whether they're diagnosed under age 50 or over age 50. The fact that we're seeing this stage shift in the wrong direction for the early-onset patient population is quite concerning for maximizing their best outcomes. This finding generated a lot of important discussions about how to reduce this difference in the stage of diagnosis for younger patients. This is important for all patients diagnosed with CRC, but because we are seeing this real difference and gap for the early-onset patient population, it warrants special attention.

Based on these findings, what actions could be taken to help combat the rise in early-onset CRC?

This [analysis] generated a lot of interesting discussion at this year’s ASCO meeting about the public health policies that should be examined or considered. First, the current age for population-wide CRC screening is 45 years in the US. That age has come down in [recent years] because of recognition that we needed to start screening patients at the population level earlier. In our analysis, half of the patients diagnosed early were diagnosed before age 45 and wouldn't have been helped by the current population-level screening guideline. That engenders the question: What should we do about screening?

For other cancers, we actually don't recommend population-wide screening, but rather targeted screening for patients who are at particularly high risk. [The patients] are most likely to benefit from the screening, and are perhaps less impacted by some of the risks or the costs of screening because of that outweighed benefit. For example, we're seeing that obesity and other metabolic diseases are likely playing a role in [the development of] early-onset CRC. [Accordingly, there could be] a strategy moving forward [we implement] a risk-based algorithm that could target certain patients with a constellation of risk factors, and obesity might be one of them. Those algorithms would need to be developed and validated, but could [provide a] rationale for screening [select] patients younger than 45.

The other important discussion point is diagnostic delay. It is well recognized in external evidence that younger patients have a longer interval of time between their first symptom and definitive diagnosis of CRC. Some studies suggest that it takes between 6 and 8 months longer for young patients [to receive a diagnosis] vs older patients, which is unacceptable. We need to reduce that diagnostic delay for those with early-onset [CRC]. I [recently] spoke with an oncologist, who suggested that patients with somewhat nonspecific gastrointestinal [GI] symptoms should be assumed to have CRC, and the threshold for referring to colonoscopy should be lowered. For these patients presenting with nonspecific GI symptoms, we should grow more comfortable as a primary care and hospitalist community referring them to colonoscopy early and often.

References

1. Carethers JM. Commencing colorectal cancer screening at age 45 years in U.S. racial groups. Front Oncol. 2022;12:966998. Published 2022 Jul 22. doi:10.3389/fonc.2022.966998
2. Herms L, Pasha S, Guo J, et al. Real-world social determinants of health (SDOH) and outcomes of early-onset colorectal cancer (EO-CRC): an analysis of a large, nationally representative US community oncology network. J Clin Oncol. 2025;43(suppl 16):11005. doi:10.1200/JCO.2025.43.16_suppl.11005