Study Finds Susceptibility Gene Variations by Race/Ethnicity in Early-Onset CRC

Andreana N. Holowatyj, PhD, MSCI

Current multigene panel tests may not be representative of early-onset colorectal cancer (CRC) risk in diverse populations, according to Andreana N. Holowatyj, PhD, MSCI.¹

A study published in the Journal of Clinical Oncology, suggests that there are racial and ethnic patterns that exist for variants in susceptibility genes APC, CHEK2, MLH1, PTEN and monoallelic MUTYH. However, no differences in overall prevalence were identified for young patients with early-onset CRC among those who are Black and White. This is true even with the pronounced disparities that exist between the 2 groups.

A total of 3,980 patients with early-onset CRC were included in the trial, and 530 germline pathogenic or likely pathogenic variants were identified in 485 patients (12.2%). Germline variants were seen in 12.7% of Ashkenazim patients, 9.5% of Asian patients, 10.3% of Black patients, 14.0% of Hispanic patients, and 12.4% of White patients.

Among patients with early-onset CRC, the prevalence of Lynch syndrome (3.7%) and specific gene variants (APC, CHEK2, MLH1, MUTYH, PTEN) differed significantly by race/ethnicity (all P <.026). Notably, Ashkenazim and Hispanic patients were more likely to have a pathogenic APC variant, including p.I1307K (OR, 2.67; 95% CI, 1.30-5.49; P =.007) , as well as a MLH1 variant (OR, 8.69; 95% CI, 2.68-28.20; P =.0003), respectively, compared with White patients.

“This is an important first step because this suggests that our current clinical multigene panel testing may not be representative of early-onset colorectal cancer risk in diverse populations,” Holowatyj, assistant professor of medicine and cancer biology at the Vanderbilt University Medical Center and Vanderbilt Ingram Cancer Center in Nashville, Tennessee, told Targeted Oncology^TM, in an interview. “I think it stands as a pillar moving forward to understand both germline and tumor changes together in a paired capacity where it will be possible to understand the implications of some of these genetic features on early-onset colorectal cancer development, but also taking a look at this on an ancestry-specific level.”

In the interview, Holowatyj discussed data from a study which found racial and ethnic differences in susceptibility genes for early-onset colorectal cancer, suggesting current multigene panel tests may not be accurate for diverse populations.

Colorectal cancer image: © mi_viri - stock.adobe.comNormal

Targeted Oncology: Can you explain the methodology for identifying germline variants and early-onset colorectal cancer across racial groups?

Holowatyj: My lab was fortunate to partner with Ambry Genetics [Corporation], a nationwide clinical testing laboratory, to identify individuals who were under the age of 50 years when they underwent germline genetic testing at Ambry Genetics who were diagnosed with a primary colorectal cancer.

What findings can discuss inherited colorectal cancer susceptibility gene variations in these patients with diverse backgrounds?

Given the pronounced disparities we know about in early-onset colorectal cancer across racial ethnic groups, which is a term based on self-reports, or self-identification, we wanted to understand whether germline or inherited genetic features could be contributing to some of the racial and ethnic disparities that we know persist in this disease burden for young patients overall. In partnering with Ambry Genetics, we were able to identify 3980 patients with early-onset colorectal cancer who underwent germline genetic testing to include in this study, and we were interested in understanding more about the prevalence and spectrum of these germline genetic features and what differences we may or may not observe across population groups.

For example, we identified that the prevalence of Lynch syndrome, which is a hereditary cancer syndrome, varied across racial ethnic groups, where we observed that about 10% of individuals who identify as Hispanic in this cohort presented with Lynch syndrome, and that went all the way down to 3% of individuals in the population who identified as Ashkenazi Jewish. In addition to differences in the prevalence of Lynch syndrome, we really wanted to understand in a diverse population what proportion of patients with early-onset colorectal cancer are presenting with at least 1 germline genetic feature that may increase susceptibility or risk for cancer development.

Overall, in the cohort, and consistent with previous literature, we observed about 12%, or 1 in every 8 patients with early-onset colorectal cancer, present with at least 1 variant or change that's associated with increased cancer susceptibility, but that number ranged from 9 and a half percent of individuals in the group who identified as Asian all the way up to 14% in the group that identified as Hispanic. In addition to looking at that broadly, we also looked at some individual gene patterns to better understand where some of these differences may lie in our study with the overall goal to understand how some of these genetic features may or may not be contributing to early-onset colorectal cancer disparities to kind of promote further discovery of potential genetic features in specific populations and groups with the goal to overall mitigate disparities and reduce the disease burden in this population.

Can you explain some of the other variations and the significance of them in the context of colorectal cancer disparities?

Previous studies have not been able to look at these variants in a diverse population. Historically, up until the point of this study, we've seen elegant work from many of my colleagues, but in cohorts of up to about 450 individuals with a large predominance of those who identify as White. The unique contribution of this study is not only being able to investigate this in nearly 4000 individuals, but we have over 1000 or so who identified as non-White. Being able to describe what this looks like in different populations is an important first step.

I think this work prompts the need for genetic discovery moving forward. Are there variants or potential changes in some of these other population groups that we've not yet characterized? We were able to study about 14 genes that all 4000 patients had uniformly assessed. These are 14 genes that are well characterized and known to have some cancer susceptibility overall, in this capacity. The next step is to broaden this work to understand this on an exome- or genome-wide level, to discover potential new variants, and also to glean a better understanding of what variants may or may not be playing a key role in some of these diverse population groups overall.

What are the potential clinical implications of these findings for across different racial groups?

This is an important first step because this suggests that our current clinical multigene panel testing may not be representative of early-onset colorectal cancer risk in diverse populations. Our study has considerations and limitations with what was available to us, but I think it stands as a pillar moving forward to understand both germline and tumor changes together in a paired capacity where it will be possible to understand the implications of some of these genetic features on early-onset colorectal cancer development, but also taking a look at this on an ancestry-specific level. As I alluded to, self-identified race and ethnicity is indeed a social construct. It's important to consider that the true biological construct is genetic ancestry. We are unable to assess that without additional molecular information.

The hope for future studies is to be able to springboard off this work, take a look at these patterns by genetic ancestry, and understand if there are new variants that we've not yet discovered. How do we need to possibly tailor some of these current multigene panel tests to ensure that they are representative of disease risk across all of these groups, and may not be well-tailored in our well-characterized population of patients who identify as non-Hispanic White, but when we see a higher disease burden in some of these populations who identify as non-White, both in disease incidence and disease characteristics, and in survival, what factors on a genetic level may we not yet know about to understand how that's contributing to this disproportionate disease burden? That is going to be key in yielding clinical implications for disease management and to improve patient outcomes overall.

How might the current multigene panel tests need to be adjusted to better assess the risk in these diverse populations?

In terms of optimizing multigene panel designs to be more representative across groups, I think that's going to predicate upon those additional studies where we can comprehensively investigate and understand what genes we should be including beyond those that are already included to ensure that it yields the optimal benefit for all patients. I don't think we're there quite yet, but I think some of the observations and seeing some unique patterns among patients who identify as Hispanic or at least among these 14 genes, seeing no differences between these germline genetic features for these genes between young White and young Black patients based on self-identification, are really important clues to help us know where to go in next steps and what we need to better study and understand.

What is the importance of tailored therapies and approaches in addressing colorectal cancer disparities?

My lab recently collaborated with a lab at Vanderbilt to publish an article on cancer discovery assessing patterns of somatic or tumor variants in early-onset colorectal cancer based on racial/ethnic groups self-identified within the AACR Project GENIE consortium. I think that study can speak to a bit more about tailoring therapeutic management from the capacity of our observation of differences in tumor mutational burden in tumors of those who identify as Black vs White. As we move towards precision prevention and therapeutics, any molecular information that we can get to better understand and move things towards targeted therapy is going to yield significant clinical implications.

One of the biggest challenges in studying this population, albeit it being incredibly important in this day and age with the rising incidence with reasons unknown, is the fact that approximately 1 in every 10 patients with colorectal cancer under age 50, so already tailored to a small population. Then, we are trying to split up this population across groups by race, ethnicity, or genetic ancestry. Those groups get even smaller, which is limiting our ability to be able to cut for larger studies that are able to yield immediate clinical implications in that capacity. This work in partnership with Ambry helped us overcome some of those existing limitations as a first step. I think the power of collaboration and institutions working together to address this important point is going to be needed for us to be able to have sufficiently sized studies and be able to tackle more in depth aspects of this important topic and question.

What other findings from this study are particularly notable?

One of the interesting observations in this study, from my perspective, was that we did not observe differences in the overall prevalence of germline genetic features, specifically between individuals who identify as White and those who identify as Black diagnosed with early-onset colorectal cancer. This raises numerous questions about what potential factors could be driving these disparities that we've observed to be different in outcomes. We know that individuals who identify as Black with early-onset colorectal cancer have poor survival compared with those who identify as White, both in colon and rectal tumors, so how do germline genetic features potentially contribute, if they do? Have we not yet identified certain ancestry specific factors? That has clinical implications.

But in addition to genetics, early-onset colorectal cancer disparities are a constellation and accomplish interplay of biology, social determinants of health, health behaviors, community factors, genetics, and biology. With our recent findings in observing differences in the tumor biology, between these 2 groups, this propels us forward to say, we need to study both pair germline and tumor sequencing data in broad and diverse population groups to understand what some of these drivers may be of tumor development, biomarkers of pathways we can clinically target, and where we can also yield some clinical implications in disease management.

No study is without its limitations. Overall, my biggest thing in terms of whenever I speak about disparities is the acknowledgement of race and ethnicity as a social construct and it being based on patient self-identification vs genetic ancestry. Hopefully weaving that piece into it in the context of telling the story is an inherent and important aspect overall.

REFERENCE:

Seagle HM, Keller SR, Tavtigian SV, Horton C, Holowatyj AN. Clinical Multigene Panel Testing Identifies Racial and Ethnic Differences in Germline Pathogenic Variants Among Patients With Early-Onset Colorectal Cancer. J Clin Oncol. 2023;41(26):4279-4289. doi:10.1200/JCO.22.02378