Pathogenic germline variants in patients with early-onset colorectal cancer according to phenotype.

doi:10.21203/rs.3.rs-5002627/v1

Download PDF

Brief Communication

Pathogenic germline variants in patients with early-onset colorectal cancer according to phenotype.

https://doi.org/10.21203/rs.3.rs-5002627/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

We assessed retrospectively the prevalence of pathogenic germline variants (PGVs) in 271 French adult patients diagnosed with colorectal cancer (CRC) before age 41, stratified by phenotype. APC, BMPR1A, CDH1, EPCAM, MLH1, MSH2, MSH3, MSH6, MUTYH, POLE, POLD1, PTEN, PMS2, SMAD4, and STK11were analysed. Overall, 21.4% of cases carried a PGV. A high prevalence was observed in Mismatch Repair-deficient (MMRd) CRC (59.7%, MMR genes) and CRC associated with polyposis (48%, APC, biallelic MUTYH and MSH3, POLE). In contrast, MMR-proficient, non-polyposis cases only had a 1.7% prevalence. The only gene involved in this third group was POLE, and two out of three cases had either two synchronous CRC or a CRC family history. PGV prevalence is dependent on CRC phenotype, questioning the need for systematic germline testing in early-onset cases. A more targeted approach, focusing on MMRd CRC, or CRC associated with polyposis, might be warranted.

Health sciences/Diseases/Cancer/Cancer genetics

Health sciences/Risk factors

The latest American National Comprehensive Cancer Network (NCCN) guidelines (2023) recommend germline genetic testing for all patients with colorectal cancer (CRC) diagnosed before the age of 50 years ¹. According to French recommendations, all patients below the age of 41 years are to be referred for germline testing ². European Society for Medical Oncology (ESMO) guidelines are more conservative, as they only advise testing in the presence of a Mismatch Repair-deficient (MMRd) CRC, in patients with polyposis, or above a certain predisposition probability threshold based on prediction models ³. Of note, ESMO guidelines were published as far back as 2019, versus 2023 for the NCCN and French ones.

The NCCN and French guidelines seem supported by recent studies showing an overall pathogenic or likely pathogenic germline variant (PGV) prevalence of about 20% among patients with early-onset CRC ^4,5.

We made the hypothesis that prevalence was actually phenotype-dependent, and that characteristics such as Mismatch Repair proficiency or absence of polyposis could obviate the need for germline testing.

We tested it by conducting a retrospective study of PGV prevalence among a large series of French CRC cases diagnosed < 41 years, grouped by phenotype. Our aim was to guide future practice, i.e., confirm systematic germline testing of early-onset cases, or allow for more conservative case selection.

We explored all cases with CRC diagnosed <41 years, referred to the adult Cancer Genetics clinics of two large Paris University Hospitals, Saint-Antoine and Cochin, between 2013 and 2023 (Saint-Antoine) and 2018 and 2023 (Cochin), and in whom germline panel testing had been performed (see below). All cases had signed a consent form for testing, and agreed to the possibility of further research based on collected data. They, or their next of kin whenever possible, were also informed of this retrospective study with a possibility to opt out. French law did not require ethics committee approval given the retrospective nature of the study.

We assigned cases to the following groups, based on CRC phenotype and associated polyps:

-Group 1. MMRd CRC as determined by immunohistochemistry

-Group 2. Polyposis, i.e., history of ≥10 colorectal polyps regardless of histology

-Group 3. MMR proficient (MMRp) CRC with no or fewer than 10 polyps

Germline testing consisted of a panel of gastrointestinal cancer susceptibility gene that included APC, BMPR1A, CDH1, EPCAM, MLH1, MSH2, MSH3, MSH6, MUTYH, POLE, POLD1, PTEN, PMS2, SMAD4, and STK11. Libraries were prepared according to the Kapa sample preparation protocol (Kapa Biosystems®). They were pooled and captured using the SeqCap EZ Choice Library (Roche/NimbleGen, Madison, WI) according to the manufacturer’s protocol and sequenced using an Illumina Platform (Illumina, San Diego, CA). PGV were confirmed by Sanger sequencing and MLPA (for copy number variation).

We included 271 cases. One hundred and forty-four were female, 127 were male. Mean age at diagnosis was 33 years (range 17-40). Fifty-eight cases carried a PGV, the overall PGV detection rate was thus 21.4% (58/271) (figure 1).

Group 1 (MMRd CRC)

A PGV was identified in 43/72 cases (59.7%). The genes involved were MLH1 (n=23), MSH2 (n=17), MSH6 (n=1), and PMS2 (n=2) (figure 2).

Group 2 (CRC and polyposis)

A PGV was identified in 12/25 cases (48%). Among the twelve carriers, six carried an APC PGV, and three MUTYH biallelic PGV. One case carried MSH3 biallelicPGV, one a POLE PGV, and one a BMPR1A PGV (figure 2). Regarding CRC-associated polyps, their number and type matched the susceptibility syndrome. All but one APC cases had >100 adenomatous polyps, while the remaining case had a “large number” of such lesions (number not specified). Biallelic MUTYH and MSH3 cases had between 16 and 50 adenomatous polyps. The POLE case had a “large number” (not specified further) of adenomatous polyps. The BMPR1A case had 30 polyps, most of them juvenile.

Group 3 (MMRp, non-polyposis CRC)

A PGV was identified in 3/174 cases (1.7%). The only gene involved in this group was POLE (figure 2). One case had two synchronous CRC aged 31 years. Another one had three 1^st- or 2^nd degree relatives with CRC before the age of 50 years. The remaining case had no other manifestation, personal or family history suggestive of POLE-associated susceptibility to CRC.

We report an overall PGV prevalence of 21% (58/271) among adult patients with CRC diagnosed <41 years. Prevalence was high in cases with either MMRd CRC (group 1) or CRC associated with polyposis (group 2): 59.7% and 48%, respectively. However, among MMRp CRC patients without associated polyposis (group 3), only 1.7% (n=3) carried a PGV. Importantly, one of them had two synchronous CRC, and another a strong family history of CRC.

These observations show that PGV prevalence differs greatly by phenotype. In the absence of MMRd or polyposis, it is very low, even more so when patients with a single isolated CRC and negative family history are considered. In this context, one wonders whether systematic germline testing is really justified. Available resources are a major determinant. State-funded and social security-based might recommend germline testing only when specific criteria are met.

Our overall results are concordant with the literature. Coughlin et al. reported a PGV prevalence of 17-26% in CRC cases <40 years tested at a commercial laboratory ⁴. In another study, among 130 unselected CRC patients diagnosed <50 years, twenty-five carried a PGV (19%) ⁵.

The strength of our study was to divide our cases in three groups, and therefore to assess PGV according to specific features (MMR status, associated polyposis).

Of note, we did not report CHEK2 PGV or the APC I1307K variant, as recent data suggest either no association with CRC or an association that too weak to be clinically relevant ^6,7.

Lynch syndrome was identified in 59.7% of cases with MMRd CRC. The remainder likely had either two somatic events in their tumor, or carried a structural variant missed by NGS ^8–10 . Complementary investigations are ongoing on a case-by-case basis.

In conclusion, overall PGV prevalence is high in adult patients with CRC diagnosed before the age of 41 years. However, this observation does not apply to MMRp CRC without associated polyposis, and these cases likely should not be offered systematic germline testing. Phenotypic features should be taken into account during genetic counseling and for testing decision.

Data availability statement

Data are available on reasonable request from European Union researchers, and pending a data transfer agreement between institutions.

Author contributions

Genetic counseling, patient management: AD, MD, JM, SF, JN, PRB.

Data collection and interpretation: AD, MD, JM, SF, PRB.

Genetic analyses: NB, AC, FC.

Manuscript writing: PRB

Final approval of manuscript: all authors

Funding

No funding was received for this study.

Ethical approval

In France, ethics committee approval is not required for retrospective studies such as this one. However, and in accordance with French law, all participants were sent an information note with the possibility to opt out.

Competing interests

PRB has received honoraria from AstraZeneca, MSD and BMS, and funding support from Astrazeneca (unrelated to the present study).

NCCN Guidelines Genetic/Familial High-Risk Assessment: Colorectal version 2.2023. Published online October 2023. Accessed July 18, 2024. https://www.nccn.org/professionals/physician_gls/pdf/genetics_colon.pdf
Lecomte T, Tougeron D, Chautard R, et al. Non-metastatic colon cancer: French Intergroup Clinical Practice Guidelines for diagnosis, treatments, and follow-up (TNCD, SNFGE, FFCD, GERCOR, UNICANCER, SFCD, SFED, SFRO, ACHBT, SFP, AFEF, and SFR). Dig Liver Dis. 2024;56(5):756-769. doi:10.1016/j.dld.2024.01.208
Stjepanovic N, Moreira L, Carneiro F, et al. Hereditary gastrointestinal cancers: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann Oncol. 2019;30(10):1558-1571. doi:10.1093/annonc/mdz233
Coughlin SE, Heald B, Clark DF, et al. Multigene Panel Testing Yields High Rates of Clinically Actionable Variants Among Patients With Colorectal Cancer. JCO Precis Oncol. 2022;6:e2200517. doi:10.1200/PO.22.00517
You YN, Moskowitz JB, Chang GJ, et al. Germline Cancer Risk Profiles of Patients With Young-Onset Colorectal Cancer: Findings From a Prospective Universal Germline Testing and Telegenetics Program. Dis Colon Rectum. 2023;66(4):531. doi:10.1097/DCR.0000000000002347
Valle L, Katz LH, Latchford A, et al. Position statement of the International Society for Gastrointestinal Hereditary Tumours (InSiGHT) on APC I1307K and cancer risk. J Med Genet. 2023;60(11):1035-1043. doi:10.1136/jmg-2022-108984
Mundt E, Mabey B, Rainville I, et al. Breast and colorectal cancer risks among over 6,000 CHEK2 pathogenic variant carriers: A comparison of missense versus truncating variants. Cancer Genet. 2023;278-279:84-90. doi:10.1016/j.cancergen.2023.10.002
Pope BJ, Clendenning M, Rosty C, et al. Germline and Tumor Sequencing as a Diagnostic Tool To Resolve Suspected Lynch Syndrome. J Mol Diagn JMD. 2021;23(3):358-371. doi:10.1016/j.jmoldx.2020.12.003
Te Paske IB, Mensenkamp AR, Neveling K, et al. Noncoding Aberrations in Mismatch Repair Genes Underlie a Substantial Part of the Missing Heritability in Lynch Syndrome. Gastroenterology. 2022;163(6):1691-1694.e7. doi:10.1053/j.gastro.2022.08.041
Bjørnstad PM, Aaløkken R, Åsheim J, et al. A 39 kb structural variant causing Lynch Syndrome detected by optical genome mapping and nanopore sequencing. Eur J Hum Genet. 2024;32(5):513-520. doi:10.1038/s41431-023-01494-7

There is no duality of interest

Download PDF

Editorial decision: revise
17 Oct, 2024
Review #2 received at journal
11 Oct, 2024
Review #1 received at journal
26 Sep, 2024
Reviewer #2 agreed at journal
25 Sep, 2024
Reviewer #1 agreed at journal
24 Sep, 2024
Reviewers invited by journal
20 Sep, 2024
Submission checks completed at journal
13 Sep, 2024
Editor assigned by journal
30 Aug, 2024
First submitted to journal
30 Aug, 2024

You are reading this latest preprint version

Pathogenic germline variants in patients with early-onset colorectal cancer according to phenotype.

Status:

Version 1

Abstract

Figures

Introduction

Material and Methods

Results

Discussion

Declarations

References

Additional Declarations

Status:

Version 1