Abstract
Background and aims:
Up to 20% of younger patients (age <50 years) diagnosed with colorectal cancer (CRC) have germline mutations in cancer susceptibility genes. Germline genetic testing may guide clinical management and facilitate earlier intervention in affected relatives. Few studies have characterized differences in genetic testing by race/ethnicity.
Methods:
We identified young adults (age 18-49 years) diagnosed with CRC between 2009 and 2017 in two health systems in Dallas, TX. We evaluated referral to genetic counseling, attendance at genetic counseling appointments, and receipt of germline genetic testing by race/ethnicity.
Results:
Of 385 patients with young-onset CRC (median age at diagnosis 44.4 years), 176 (45.7%) were Hispanic, 98 (25.4%) non-Hispanic Black, and 111 (28.8%) non-Hispanic White. Most patients (76.9%) received immunohistochemistry (IHC) for mismatch repair proteins, and there was no difference in receipt of IHC by race/ethnicity. However, a lower proportion of Black patients were referred to genetic counseling (50.0% vs. White patients 54.1% vs. Hispanic patients 65.9%, p=0.02) and attended genetic counseling appointments (61.2% vs. 81.7% White patients vs. 86.2% Hispanic patients, p<0.01). Of 141 patients receiving genetic testing, 38 (27.0%) had a pathogenic or likely pathogenic variant in a cancer susceptibility gene. An additional 33 patients (23.4%) had variants of uncertain significance, of which 84.8% occurred in racial/ethnic minorities.
Conclusion:
In a diverse population of patients diagnosed with young-onset CRC, we observed racial/ethnic differences in referral to and receipt of germline genetic testing. Our findings underscore the importance of universal genetic testing to address racial/ethnic disparities in young-onset CRC.
Keywords: colorectal cancer, hereditary syndrome, comparison, race, ethnicity
Introduction
Colorectal cancer (CRC) is the third most common cancer in the United States, and a growing number of new diagnoses occur in persons under age 50 years.1, 2 Incidence rates of CRC in young adults (age 18-49 years) increased from 9.2 per 100,000 in 2000 to 12.9 per 100,000 in 2017, with the largest absolute increases among 40-49 year olds.3, 4 Mortality rates increased from 2.7 per 100,000 to 3.1 per 100,000 during the same period. Young-onset CRC disproportionately affects racial and ethnic minorities. Although 10-15% of all patients diagnosed with CRC are younger than age 50 years, the proportion is nearly doubled among non-Hispanic Black patients (16%) compared to White patients (9%).5, 6 Incidence rates are also rapidly increasing among Hispanic persons.7-9
Hereditary cancer syndromes are relatively common among patients diagnosed with young-onset CRC.10 Advances in next-generation sequencing (NGS) have made germline genetic testing more accessible and cost-effective.11, 12 As many as 20% of young-onset CRC diagnoses are associated with hereditary cancer syndromes or pathogenic mutations in cancer susceptibility genes.13-15 However, many patients do not have tumor phenotypes typically associated with these hereditary syndromes, and others harbor pathogenic mutations without a family history of CRC or cancer.14, 15 Universal genetic testing may: 1) identify patients with hereditary syndromes not meeting traditional clinical criteria; 2) guide management of those with clinically actionable mutations; and 3) facilitate cascade testing and earlier intervention in affected relatives.
Information on genetic testing in minority populations is limited. To address this gap, we characterized racial/ethnic differences in receipt of germline genetic testing, including referral to and attendance at genetic counseling appointments, in a diverse population of patients with young-onset CRC. We also explored differences in prevalence of pathogenic mutations across racial/ethnic groups.
Materials and Methods
We identified young adults (age 18-49 years) diagnosed with or treated for histologically-confirmed colorectal adenocarcinoma at UT Southwestern Medical Center (UT Southwestern) and Parkland Health and Hospital System (Parkland) between January 1, 2009 and June 30, 2017.16 These health systems serve patients of different sociodemographic backgrounds. UT Southwestern is an academic tertiary-care referral center and primarily cares for patients with private insurance or Medicare. As a safety-net health system, Parkland is the sole provider of cancer care for patients without insurance in Dallas County. Both health systems have integrated electronic medical records (EMR) for all inpatient and outpatient care and maintain tumor registries that meet quality standards of the Commission on Cancer and National Program of Cancer Registries.17 Genetic counselors at UT Southwestern serve several clinical sites in North Texas, including Parkland.18 This study was approved by the Institutional Review Board at UT Southwestern Medical Center (#022017-038).
Patients were identified using the hospitals’ tumor registries, and we then manually reviewed the EMR for patient demographics, family history of CRC or Lynch-related malignancies, tumor characteristics, and receipt of treatment (surgery, chemotherapy, radiation therapy). Tumor characteristics included histology, histologic grade, and mismatch repair deficiency. For most of the study period (2010-2017), both health systems used a universal screening protocol for Lynch syndrome for patients diagnosed with CRC and younger than age 70 years.19 The protocol included immunohistochemistry (IHC) for the presence of one of the four DNA mismatch repair proteins (MLH1, MSH2, MSH6, PMS2). Tumors were classified as mismatch repair protein deficient (dMMR) if there was >90% loss of one or more proteins.
We examined three outcomes: 1) referral to genetic counseling, defined as the proportion of all patients for whom a referral to genetic counseling was placed; 2) attendance at genetic counseling appointments, defined as the proportion of patients referred to genetic counseling who attended the appointment; and 3) receipt of germline genetic testing, defined as the proportion of patients attending genetic counseling appointments who completed genetic testing. We considered patients to have been referred to genetic counseling if a provider placed a referral to the genetics clinic in the EMR. For patients attending genetic counseling appointments, we also examined receipt of gene-specific vs. multigene panel testing and test results.
Statistical Analysis
We compared demographics, tumor characteristics, and the three outcomes (i.e., referral, attendance, receipt) by race/ethnicity using Chi-square or Fisher’s exact tests, as appropriate. We also describe IHC and germline genetic testing results by race/ethnicity. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC), and statistical significance was defined as p<0.05 for all analyses.
Results
Characteristics of 409 patients with young-onset CRC at UT Southwestern (n=94) and Parkland (n=315) are shown in Table 1. Most patients were diagnosed at age 40 -49 years (71.6%); median age at diagnosis was 44.4 years (IQR 39.3-47.6 years). The cohort was racially and ethnically diverse: 27.1% were non-Hispanic White, 24.0% non-Hispanic Black, 43.0% Hispanic, and 5.9% Asian. Anatomic subsite (approximately 45% distal colon) and stage (approximately 75% stage III-IV) did not statistically significantly differ across racial/ethnic groups. Due to small numbers, we excluded Asians (n=24) from analyses of IHC and genetic testing.
Table 1.
Characteristics of 409 patients with young-onset colorectal cancer (age 18-49 years), UT Southwestern Medical Center and Parkland Health & Hospital System, 2009 – 2017, by race/ethnicity
| Non-Hispanic White (n=111) |
Non-Hispanic Black (n=98) |
Hispanic (n=176) |
Asian (n=24) |
||||||
|---|---|---|---|---|---|---|---|---|---|
| n | % | N | % | n | % | n | % | p-value | |
| Age, median (IQR) | 44.9 (40.4 - 47.2) | 45.9 (39.8 - 48.2) | 43.3 (37.9 - 47.4) | 44.2 (40.0-46.7) | 0.09 | ||||
| Sex | 0.89 | ||||||||
| Male | 63 | 56.8 | 53 | 54.1 | 97 | 55.1 | 15 | 62.5 | |
| Female | 48 | 43.2 | 45 | 45.9 | 79 | 44.8 | 9 | 37.5 | |
| Family history CRC1 | 0.59 | ||||||||
| No family history | 75 | 67.6 | 69 | 70.4 | 134 | 76.7 | 17 | 70.8 | |
| Any family history | 32 | 28.8 | 28 | 28.6 | 42 | 23.3 | 7 | 29.2 | |
| First-degree relative | 7 | 6.3 | 15 | 15.3 | 21 | 11.9 | 4 | 16.7 | |
| Second-degree relative | 24 | 21.6 | 16 | 16.3 | 24 | 13.6 | 4 | 16.7 | |
| Third-degree relative | 4 | 3.6 | 4 | 4.1 | 5 | 2.8 | - | - | |
| Unknown2 | 4 | 3.6 | 1 | 1.0 | - | - | - | - | |
| Anatomic subsite3 | 0.1 | ||||||||
| Proximal colon | 21 | 18.9 | 32 | 32.6 | 50 | 28.4 | 8 | 33.3 | |
| Distal colon | 50 | 45.0 | 46 | 46.9 | 76 | 43.2 | 11 | 45.8 | |
| Rectum | 40 | 36.0 | 20 | 20.4 | 50 | 28.4 | 5 | 20.8 | |
| Stage at diagnosis | 0.24 | ||||||||
| In situ, 0, or I | 11 | 9.9 | 10 | 10.2 | 12 | 6.8 | 4 | 16.7 | |
| II | 15 | 13.5 | 14 | 14.3 | 34 | 19.3 | 6 | 25.0 | |
| III | 51 | 45.9 | 32 | 32.3 | 69 | 39.2 | 9 | 37.5 | |
| IV | 34 | 30.6 | 42 | 42.9 | 61 | 34.7 | 5 | 20.8 | |
| Histology | 0.15 | ||||||||
| Adenocarcinoma | 99 | 89.2 | 90 | 91.8 | 153 | 86.9 | 18 | 75.0 | |
| Mucinous or signet ring | 12 | 10.8 | 8 | 8.2 | 23 | 13.1 | 6 | 25.0 | |
| Histologic grade | 0.26 | ||||||||
| Well/moderately differentiated | 85 | 76.6 | 67 | 68.3 | 129 | 73.3 | 14 | 58.3 | |
| Poorly/undifferentiated | 18 | 16.2 | 21 | 21.4 | 38 | 21.6 | 8 | 33.3 | |
| Unknown2 | 8 | 7.2 | 10 | 10.2 | 9 | 5.1 | 2 | 8.3 | |
| Received surgery | 91 | 82.0 | 80 | 81.6 | 144 | 81.8 | 19 | 79.2 | 0.51 |
| Received chemotherapy | 82 | 73.9 | 73 | 74.5 | 139 | 79.0 | 18 | 75.0 | 0.74 |
| Received radiation | 37 | 33.3 | 25 | 25.5 | 47 | 26.7 | 5 | 20.8 | <0.01 |
p-value compares any family history vs. none
Missing values not included in Chi-square or Fisher’s exact tests
Proximal colon includes cecum up to hepatic flexure; distal colon includes splenic flexure up to rectosigmoid junction
Racial/ethnic differences in receipt and results of IHC testing
About three-fourths of patients (n=296 of 385; 76.9%) received IHC for MMR protein deficiency (Figure 1), with no significant difference in receipt of IHC by race/ethnicity (p=0.48, Table 2). A higher proportion of Hispanic patients had dMMR tumors compared to White and Black patients (23.4% vs. 8.6% and 7.7% respectively, p<0.01).
Figure 1.
Study flow diagram showing receipt and results of IHC and germline genetic testing (n=385)
Table 2.
Receipt of IHC testing and mismatch repair protein deficiency, overall and by race/ethnicity (n=385)
| All patients (n=385) |
Non-Hispanic White (n=111) |
Non-Hispanic Black (n=98) |
Hispanic (n=176) |
p-value | |
|---|---|---|---|---|---|
| Received IHC testing | 296 (76.8) | 81 (73.0) | 78 (79.5) | 137 (77.8) | 0.48 |
| MMR protein deficient | 45/296 (15.2) | 7/81 (8.6) | 6/78 (7.7) | 32/137 (23.4) | <0.01 |
| IHC results | |||||
| MLH1/PMS2 loss | 24/45 (55.3) | 3/7 (42.9) | 0/6 (0) | 21/32 (65.6) | |
| MLH1 loss | 1/45 (2.2) | 1/7 (14.3) | 0/6 (0) | 0/32 (0) | |
| PMS2 loss | 3/45 (6.7) | 0/7 (0) | 1/6 (16.7) | 2/32 (6.3) | |
| MSH2/MSH6 loss | 10/45 (22.2) | 1/7 (14.3) | 4/6 (66.7) | 5/32 (15.6) | |
| MSH2 loss | 2/45 (4.4) | 0/7 (0) | 1/6 (16.7) | 1/32 (3.1) | |
| MSH6 loss | 5/45 (11.1) | 2/7 (28.6) | 0/6 (0) | 3/32 (9.4) | |
| Received reflex testing1 | 8/25 (32.0) | 1/4 (25.0) | -- | 7/21 (33.3) |
NOTE: Due to small numbers, Asians (n=24) were excluded from this table
Reflex testing for sporadic MLH1 loss with MLH1 hypermethylation or BRAF V600E mutation testing in patients who had MLH1 deficiency
Racial/ethnic differences in referral to genetic counseling, attendance at genetic counseling appointments, and receipt of germline genetic testing
Over half (n=225 of 385, 58.4%) of all patients were referred for genetic counseling including almost all patients with dMMR tumors (n=41 of 43, 95.3%). However, a lower proportion of Black patients (n=49, 50.0%) and a higher proportion of Hispanic patients (n=116, 65.9%) were referred compared to White patients (n=60, 54.1%; p=0.02). This pattern was consistent across age groups (age 18-29, 30-39, and 40-49 years; Figure 2) and health systems (data not shown).
Figure 2.
Proportion of patients referred to genetic counseling by race/ethnicity and age (n=225)
Among those referred (n=225), 79.6% attended the genetic counseling appointment (Table 3). A lower proportion of Black patients (n=30, 61.2%) attended the appointment compared to White (n=49, 81.7%) and Hispanic (n=100, 86.2%) patients (p<0.01). Reasons for not attending the appointment (n=46) included: no show (n=29, 63.0%), never scheduled (n=13, 28.2%), patient died before appointment (n=2, 4.3%), and appointment scheduled in future (n=2, 4.3%).
Table 3.
Referral to genetic counseling, attendance at genetic counseling appointments, and receipt of germline genetic testing, overall and by race/ethnicity (n=385)
| All patients (n=385) |
Non-Hispanic White (n=111) |
Non-Hispanic Black (n=98) |
Hispanic (n=176) |
p-value | |
|---|---|---|---|---|---|
| Referred for genetic counseling1 | 225/385 (58.4) | 60/111 (54.1) | 49/98 (50.0) | 116/176 (65.9) | 0.02 |
| Attended genetic counseling appointment2 | 179/225 (79.6) | 49/60 (81.7) | 30/49 (61.2) | 100/116 (86.2) | <0.01 |
| Received germline genetic testing3 | 141/179 (78.8) | 37/49 (75.5) | 21/30 (70.0) | 83/100 (83.0) | 0.25 |
| Gene-specific testing | 57/141 (40.4) | 14/37 (37.8) | 8/21 (38.1) | 35/83 (42.2) | |
| Multigene panel testing | 84/141 (59.6) | 23/37 (62.2) | 13/21 (61.9) | 48/83 (57.8) | |
| Pathogenic or likely pathogenic variant in cancer susceptibility gene | 38/141 (27.0) | 7/37 (18.9) | 7/21 (33.3) | 24/83 (28.9) | 0.41 |
| Lynch syndrome (MLH1, PMS2, MSH2, MSH6) | 20/141 (14.2) | 3/37 (8.1) | 2/21 (9.5) | 15/83 (18.1) | |
| Polyposis syndromes (APC, MYH, BMPR1A) | 12/141 (8.5) | 2/37 (5.4) | 3/21 (14.3) | 7/83 (8.4) | |
| Other susceptibility genes | 6/141 (4.3) | 2/37 (5.4) | 2/21 (9.5) | 2/83 (2.4) | |
| Variant of uncertain significance | 33/141 (23.4) | 5/37 (13.5) | 3/21 (14.3) | 25/83 (30.1) | 0.10 |
NOTE: Due to small numbers, Asians (n=24) were excluded from this table
Referral to genetic counseling defined as the proportion of all patients for whom a referral to genetic counseling was placed
Attendance at genetic counseling appointments defined as the proportion of patients referred to genetic counseling who attended the appointment
Receipt of germline genetic testing defined as the proportion of patients attending genetic counseling appointments who completed genetic testing
Most patients (n=141, 78.8%) who attended genetic counseling appointments received germline genetic testing (Table 3). Receipt of genetic testing did not statistically significantly differ by race/ethnicity (p=0.25), although fewer Black patients (n=21, 70.0%) and more Hispanic patients (n=83, 83.0%) received testing compared to White patients (n=37, 75.5%). Reasons for not receiving testing (n=38) included: insurance/could not afford (n=11, 28.9%), not recommended by genetic counselor/did not meet testing criteria (n=11, 28.9%), did not return saliva sample (n=6, 15.8%), patient declined (n=7, 18.4%), and lab error (n=3, 7.9%). Reasons for no receipt of testing did not appear to differ by race/ethnicity.
Among those tested (n=141), over half (n=84 of 141, 59.6%) received multigene panel testing, and others (n=57 of 141, 40.4%) received gene-specific testing (e.g., APC or MYH gene testing for polyposis syndromes, MLH1, PMS2, MSH2, or MSH6 gene testing for Lynch syndrome) (Table 3). Most patients who received gene-specific testing were diagnosed before 2014, and thereafter multigene panels became the predominant testing modality (Supplementary Figure 1).
Nearly one-third (n=38/141, 27.0%) of patients had a pathogenic or likely pathogenic mutation in a cancer susceptibility gene (Table 3), representing 36 unique mutations (Supplementary Table 1). Specifically, 20 patients (14.2%) had Lynch syndrome, 12 (8.5%) had a polyposis syndrome, and 6 (4.3%) had pathogenic or likely pathogenic variants in other susceptibility genes (e.g., TP53, CHEK2, BRCA2) (Table 4). A higher proportion of Hispanic patients had Lynch syndrome compared to other groups (18.1% vs. 8.1% White patients vs. 9.5% Black patients). Hispanic patients with Lynch syndrome most frequently had pathogenic mutations in MLH1 or MSH2, and some of these mutations were observed more than once (Supplementary Table 1). For example, two and three patients (not known to be related) had the same pathogenic mutations in MLH1 – c.1168_1169insG and c.1790_1791del2ins9, respectively.
Table 4.
Pathogenic mutations and associated syndromes, reported as proportion of patients receiving germline genetic testing, overall and by race/ethnicity (n=141)
| Gene | Hereditary syndrome or associated cancer(s) |
All patients (n= 441) |
Non-Hispanic white (n =37) |
Non-Hispanic black (n =21) |
Hispanic (n =83) |
||||
|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | ||
| Any pathogenic variant | 38 | 30.0 | 7 | 18.9 | 7 | 33.3 | 24 | 28.9 | |
| Genes associated with CRC | |||||||||
| MLH1 | Lynch syndrome | 10 | 7.1 | 1 | 2.7 | - | - | 9 | 10.1 |
| MSH2 | Lynch syndrome | 6 | 4.2 | 1 | 2.7 | 1 | 4.8 | 4 | 4.8 |
| PMS2 | Lynch syndrome | 2 | 1.4 | - | - | 1 | 4.8 | 1 | 1.2 |
| MSH6 | Lynch syndrome | 2 | 1.4 | 1 | 2.7 | - | - | 1 | 1.2 |
| APC | Familial adenomatous polyposis | 8 | 5.7 | 1 | 2.7 | 3 | 14.3 | 4 | 4.8 |
| MYH | MUTYH-associated polyposis biallelic | 3 | 2.1 | 1 | 2.7 | - | - | 2 | 2.4 |
| BMPR1A | Juvenile polyposis syndrome | 1 | 0.7 | - | - | - | - | 1 | 1.2 |
| TP53 | Li-Fraumeni syndrome | 1 | 0.7 | - | - | 1 | 4.8 | - | - |
| Genes not typically associated with CRC | |||||||||
| ATM | Ataxia-telangiectasia syndrome | 1 | 0.7 | - | - | - | - | 1 | 1.2 |
| BRCA1 | Hereditary breast-ovarian cancer syndrome | 1 | 0.7 | - | - | - | - | 1 | 1.2 |
| BRCA2 | Hereditary breast-ovarian cancer syndrome | 1 | 0.7 | 1 | 2.7 | - | - | - | - |
| CHEK2 | Breast cancer, pancreatic cancer | 2 | 1.4 | 1 | 2.7 | 1 | 4.8 | - | - |
NOTE: Due to small numbers, Asians (n=24) were excluded from this table
Thirty-three patients (23.4% of those tested) had variants of uncertain significance (VUS) (Table 3), including four patients who also had a pathogenic mutation. The majority (n=28, 84.8%) of VUS were identified in racial/ethnic minorities. Specifically, we observed a higher proportion of Hispanic (30.1%) patients with VUS compared to White patients (13.5%), although robust analyses were limited by small numbers (p=0.10).
Discussion
In a racially and ethnically diverse population, we found over three-fourths of patients diagnosed with young-onset CRC received tumor screening with IHC for Lynch syndrome, and over half were referred for germline genetic testing. However, referral to and attendance at genetic counseling appointments differed by race/ethnicity. Black patients were significantly less likely to be referred to genetic counseling, and less likely to attend genetic counseling appointments, compared to Hispanic and White patients. Among those tested, nearly one-third of patients had a pathogenic or likely pathogenic variant in a cancer susceptibility gene. New diagnoses of Lynch syndrome were common in Hispanic patients, among whom mutations in MLH1 were observed repeatedly. Similarly, a higher-than-expected proportion of racial/ethnic minorities had VUS.
Despite increasing availability of germline genetic testing, our results suggest genetic testing is a complex process comprised of several steps: accurate patient recall and provider assessment of family history of cancer, provider order of relevant clinical tests (e.g., IHC), provider referral to genetic counseling, patient attendance at genetic counseling appointment, and if recommended, patient completion or return of testing. Barriers to completing each of these steps persist, particularly in minority populations.20 For example, fewer Black patients in our study were referred to and attended genetic counseling appointments compared to Hispanic and White patients. Racial/ethnic differences in referral persisted across both health systems, suggesting these differences are not explained by differences in care delivery between the two systems. Prior studies of colorectal, breast, and ovarian cancer similarly report patients are not routinely referred to genetic counseling,21 and referral varies by race/ethnicity.22-24 Differential referral may be related to several provider-level factors: implicit bias, lack of knowledge or training, uncertainty related to the importance of or how results may be used, and different referral thresholds or criteria applied across patient groups. Implicit racial/ethnic bias of healthcare providers is associated with shorter patient-provider interactions, poor patient-centered communication, and patient mistrust in and adherence to clinical recommendations.25, 26 Patient-level factors, such as competing medical needs, mistrust in the health system, cultural stigmas related to cancer and familial syndromes,22, 27, 28 or other social inequities29 may also lead some patients to never attend genetic counseling appointments and play a role in the disparities we observed. Many of these factors can be traced back to patient experiences of racism30 or may be a result of structural racism.31
Among those attending genetic counseling appointments, there was no difference in receipt of germline genetic testing by race/ethnicity, suggesting disparities may be eliminated by intervening on earlier steps in the testing process. For example, implementing automated referrals by gastroenterologists at the time of diagnosis may increase the proportion of patients referred to genetic counseling, equally across racial/ethnic groups. Similarly, because many patients are diagnosed by gastroenterologists, education and training activities tailored for gastroenterologists may be a complementary strategy to eliminate disparities. Alternative models of care delivery, such as telemedicine or telephone counseling, may also remove logistic barriers to attending appointments, or reduce the burden of clinic visits required while patients receive intensive cancer therapy. Several studies show patients are receptive to telemedicine and report high levels of satisfaction receiving genetic counseling services through this model.32
Genetic testing for hereditary cancer syndromes has historically been limited to patients meeting clinical criteria for testing, such as Amsterdam or Bethesda criteria to identify those with Lynch syndrome, but these stringent criteria can miss a substantial proportion of patients with hereditary syndromes. Reports from the Ohio Colorectal Cancer Prevention Initiative13 and the University of Michigan Comprehensive Cancer Center15 suggest as many as 20% of patients diagnosed with young-onset CRC have Lynch syndrome. Our study adds to this literature by highlighting racial/ethnic differences in MMR phenotypes and pathogenic mutations in Lynch-related genes. Double the proportion of Hispanic patients had Lynch syndrome in our study compared to White and Black patients and of that reported elsewhere (range of 4-14%).13, 15 A higher proportion of Hispanic patients also had dMMR tumors. Although differences in prevalence of Lynch syndrome did not reach statistical significance, these findings generate several hypotheses to explore in future studies, such as the penetrance and expression of Lynch-related genes in Hispanic families. The majority of Hispanic patients receiving care at our two health systems originate from northern Mexico, and this patient population may be enriched for founder mutations that have not been previously appreciated. Specifically, we observed two mutations in MLH1 repeatedly in Hispanic patients not known to be related. To our knowledge, c.1790_1791del2ins9 has only been reported once in a study of Hispanic patients with Lynch syndrome in California, Texas, and Puerto Rico,33 and c.1168_1169insG has not yet been described in the literature. Given the rapidly increasing incidence of young-onset CRC in Hispanic persons,9 scarcity of data regarding Lynch syndrome in this population, and continued advances in NGS, the prevalence and characteristics of Lynch syndrome in Hispanic patients requires additional study.
Similarly, a notable proportion of Hispanic patients in our study had variants of uncertain significance compared to White patients. Because most studies of hereditary cancer syndromes are conducted in European populations, and racial/ethnic minorities are often underrepresented in reference databases used for variant annotation, clinical criteria to identify hereditary syndromes may be less accurate in minority populations, leading to uncertain results or unrecognized genetic risk factors.34-36 We expected prevalence of VUS to differ by race/ethnicity, but we were surprised that almost 90% of VUS occurred in non-White patients. VUS pose many challenges, including a lack of guidance for assigning risk to carriers, difficulty in communicating results to patients, reclassifying variants, and re-contacting patients with reclassified variants. Several studies report discomfort among physicians, and even genetic counselors, in relaying information about VUS to patients, especially without guidance from the reporting laboratory or institution.37, 38 Other studies have demonstrated that patients with VUS experience more cancer-related distress compared to those with negative results.39 Racial/ethnic minorities may be particularly vulnerable to these challenges because of access to care, language barriers, and health literacy. Expanding the reach of genetic testing to underserved and minority populations may enable more rapid reclassification of VUS.
Our study had several limitations. Although we identified important racial/ethnic disparities in the genetic testing process, including referral to genetic counseling and attendance at appointments, the retrospective design of our study precluded us from identifying reasons for these disparities. The small number of patients with pathogenic and likely pathogenic mutations limited statistical power for some comparisons, particularly evaluating differences in prevalence of specific hereditary syndromes. Not all patients in our study received tumor screening with IHC for Lynch syndrome or multigene panel testing for germline mutations, and we may have underestimated the true burden of hereditary cancer syndromes. Most Hispanic patients in our study were Hispanic White patients from Mexico, and results may not represent Hispanic patients from other regions.
In summary, all patients diagnosed with young-onset CRC should be considered for germline genetic testing because as many as 20% of patients have hereditary cancer syndromes,40, 41 and many of these patients do not meet traditional clinical criteria for testing. Genetic testing may also guide management of those with clinically actionable mutations and facilitate earlier intervention in affected relatives, most of whom may be unaware of their diagnosis. Yet, despite the promise and increasing availability of genetic testing, we observed multiple failures in the genetic testing process, particularly referral to testing, and these failures disproportionately affected racial/ethnic minorities. Our findings suggest differences in utilization of testing may contribute to observed disparities in young-onset CRC.42 A remaining and critical need is to address provider- and patient-level factors contributing to failures at each step in the genetic testing process.
Supplementary Material
What You Need to Know.
Background and Context
Up to 20% of younger patients (age <50 years) diagnosed with colorectal cancer (CRC) have germline mutations in cancer susceptibility genes. Few studies have characterized differences in genetic testing and susceptibility by race/ethnicity.
New Findings
In a racially and ethnically diverse population, Black patients were significantly less likely to be referred to genetic counseling, and less likely to attend genetic counseling appointments, compared to Hispanic and White patients. Among those tested, nearly one-third of patients had a pathogenic or likely pathogenic variant in a cancer susceptibility gene. New diagnoses of Lynch syndrome were common in Hispanic patients, among whom mutations in MLH1 were observed repeatedly.
Limitation
The retrospective design of our study precluded us from identifying reasons for racial/ethnic disparities in referral to and attendance at genetic counseling appointments.
Impact
Our findings suggest differences in utilization of germline genetic testing may contribute to observed disparities in young-onset colorectal cancer.
Abbreviations:
- CRC
colorectal cancer
- yoCRC
young-onset colorectal cancer
- NGS
next generation sequencing
- IHC
immunohistochemistry
- MMR
mismatch repair
- EMR
electronic medical record
Footnotes
Conflict of Interest: AS has served as a consultant for Exact Sciences. None of the other authors have any relevant conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Siegel RL, Miller KD, Fedewa SA, et al. Colorectal cancer statistics, 2017. CA Cancer J Clin 2017;67:177–193. [DOI] [PubMed] [Google Scholar]
- 2.Siegel RL, Fedewa SA, Anderson WF, et al. Colorectal Cancer Incidence Patterns in the United States, 1974-2013. J Natl Cancer Inst 2017;109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bailey CE, Hu CY, You YN, et al. Increasing disparities in the age-related incidences of colon and rectal cancers in the United States, 1975-2010. JAMA Surg 2015;150:17–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Stat S.
- 5.Stoffel EM, Murphy CC. Epidemiology and Mechanisms of the Increasing Incidence of Colon and Rectal Cancers in Young Adults. Gastroenterology 2020;158:341–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Murphy CC, Sanoff HK, Stitzenberg KB, et al. Patterns of Sociodemographic and Clinicopathologic Characteristics of Stages II and III Colorectal Cancer Patients by Age: Examining Potential Mechanisms of Young-Onset Disease. J Cancer Epidemiol 2017;2017:4024580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rahman R, Schmaltz C, Jackson CS, et al. Increased risk for colorectal cancer under age 50 in racial and ethnic minorities living in the United States. Cancer Med 2015;4:1863–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stefanidis D, Pollock BH, Miranda J, et al. Colorectal cancer in Hispanics: a population at risk for earlier onset, advanced disease, and decreased survival. Am J Clin Oncol 2006;29:123–6. [DOI] [PubMed] [Google Scholar]
- 9.Garcia S, Pruitt SL, Singal AG, et al. Colorectal cancer incidence among Hispanics and non-Hispanic Whites in the United States. Cancer Causes Control 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol 2008;26:5783–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gallego CJ, Shirts BH, Bennette CS, et al. Next-Generation Sequencing Panels for the Diagnosis of Colorectal Cancer and Polyposis Syndromes: A Cost-Effectiveness Analysis. J Clin Oncol 2015;33:2084–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tan O, Shrestha R, Cunich M, et al. Application of next-generation sequencing to improve cancer management: A review of the clinical effectiveness and cost-effectiveness. Clin Genet 2018;93:533–544. [DOI] [PubMed] [Google Scholar]
- 13.Pearlman R, Frankel WL, Swanson B, et al. Prevalence and Spectrum of Germline Cancer Susceptibility Gene Mutations Among Patients With Early-Onset Colorectal Cancer. JAMA Oncol 2017;3:464–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yurgelun MB, Kulke MH, Fuchs CS, et al. Cancer Susceptibility Gene Mutations in Individuals With Colorectal Cancer. J Clin Oncol 2017;35:1086–1095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Stoffel EM, Koeppe E, Everett J, et al. Germline Genetic Features of Young Individuals With Colorectal Cancer. Gastroenterology 2018;154:897–905.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dharwadkar P, Greenan G, Singal AG, et al. Is Colorectal Cancer in Patients Younger Than 50 Years of Age the Same Disease as in Older Patients? Clin Gastroenterol Hepatol 2019. [DOI] [PubMed] [Google Scholar]
- 17.Lodrigues W, Dumas J, Rao M, et al. Compliance with the commission on cancer quality of breast cancer care measures: self-evaluation advised. Breast J 2011;17:167–71. [DOI] [PubMed] [Google Scholar]
- 18.Pritzlaff M, Yorczyk A, Robinson LS, et al. An internal performance assessment of CancerGene Connect: an electronic tool to streamline, measure and improve the genetic counseling process. J Genet Couns 2014;23:1034–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Marquez E, Geng Z, Pass S, et al. Implementation of routine screening for Lynch syndrome in university and safety-net health system settings: successes and challenges. Genet Med 2013;15:925–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hann KEJ, Freeman M, Fraser L, et al. Awareness, knowledge, perceptions, and attitudes towards genetic testing for cancer risk among ethnic minority groups: a systematic review. BMC Public Health 2017;17:503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sanchez JA, Vogel JD, Kalady MF, et al. Identifying Lynch syndrome: we are all responsible. Dis Colon Rectum 2008;51:1750–6. [DOI] [PubMed] [Google Scholar]
- 22.Muller C, Lee SM, Barge W, et al. Low Referral Rate for Genetic Testing in Racially and Ethnically Diverse Patients Despite Universal Colorectal Cancer Screening. Clin Gastroenterol Hepatol 2018;16:1911–1918.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Petzel SV, Vogel RI, Bensend T, et al. Genetic risk assessment for women with epithelial ovarian cancer: referral patterns and outcomes in a university gynecologic oncology clinic. J Genet Couns 2013;22:662–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Manrriquez E, Chapman JS, Mak J, et al. Disparities in genetics assessment for women with ovarian cancer: Can we do better? Gynecol Oncol 2018;149:84–88. [DOI] [PubMed] [Google Scholar]
- 25.Penner LA, Dovidio JF, Gonzalez R, et al. The Effects of Oncologist Implicit Racial Bias in Racially Discordant Oncology Interactions. J Clin Oncol 2016;34:2874–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hall WJ, Chapman MV, Lee KM, et al. Implicit Racial/Ethnic Bias Among Health Care Professionals and Its Influence on Health Care Outcomes: A Systematic Review. Am J Public Health 2015;105:e60–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Allford A, Qureshi N, Barwell J, et al. What hinders minority ethnic access to cancer genetics services and what may help? Eur J Hum Genet 2014;22:866–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Halbert CH, Harrison BW. Genetic counseling among minority populations in the era of precision medicine. Am J Med Genet C Semin Med Genet 2018;178:68–74. [DOI] [PubMed] [Google Scholar]
- 29.Alcaraz KI, Wiedt TL, Daniels EC, et al. Understanding and addressing social determinants to advance cancer health equity in the United States: A blueprint for practice, research, and policy. CA Cancer J Clin 2020;70:31–46. [DOI] [PubMed] [Google Scholar]
- 30.Shariff-Marco S, Klassen AC, Bowie JV. Racial/ethnic differences in self-reported racism and its association with cancer-related health behaviors. Am J Public Health 2010;100:364–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Williams DR, Lawrence JA, Davis BA. Racism and health: evidence and needed research. Annual review of public health 2019;40:105–125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hilgart JS, Hayward JA, Coles B, et al. Telegenetics: a systematic review of telemedicine in genetics services. Genetics in Medicine 2012;14:765–776. [DOI] [PubMed] [Google Scholar]
- 33.Sunga AY, Ricker C, Espenschied CR, et al. Spectrum of mismatch repair gene mutations and clinical presentation of Hispanic individuals with Lynch syndrome. Cancer Genet 2017;212–213:1-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ricker C, Culver JO, Lowstuter K, et al. Increased yield of actionable mutations using multi-gene panels to assess hereditary cancer susceptibility in an ethnically diverse clinical cohort. Cancer Genet 2016;209:130–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Guindalini RS, Win AK, Gulden C, et al. Mutation spectrum and risk of colorectal cancer in African American families with Lynch syndrome. Gastroenterology 2015;149:1446–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Moreno-Ortiz JM, Ayala-Madrigal MeL, Corona-Rivera JR, et al. Novel Mutations in MLH1 and MSH2 Genes in Mexican Patients with Lynch Syndrome. Gastroenterol Res Pract 2016;2016:5278024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Eccles BK, Copson E, Maishman T, et al. Understanding of BRCA VUS genetic results by breast cancer specialists. BMC Cancer 2015;15:936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Scherr CL, Lindor NM, Malo TL, et al. Genetic counselors' practices and confidence regarding variant of uncertain significance results and reclassification from BRCA testing. Clin Genet 2015;88:523–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Culver JO, Brinkerhoff CD, Clague J, et al. Variants of uncertain significance in BRCA testing: evaluation of surgical decisions, risk perception, and cancer distress. Clin Genet 2013;84:464–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Stoffel EM, Koeppe E, Everett J, et al. Germline Genetic Features of Young Individuals With Colorectal Cancer. Gastroenterology 2018;154:897–905.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Pearlman R, Frankel WL, Swanson B, et al. Prevalence and Spectrum of Germline Cancer Susceptibility Gene Mutations Among Patients With Early-Onset Colorectal Cancer. JAMA Oncol 2017;3:464–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Murphy CC, Wallace K, Sandler RS, et al. Racial Disparities in Incidence of Young-Onset Colorectal Cancer and Patient Survival. Gastroenterology 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.