Clinical and molecular characteristics of constitutional mismatch repair deficiency syndrome: a case series of five children and appraisal of diagnostic guidelines

Abstract

DNA mismatch repair (MMR) is critical for maintaining genome integrity through correction of single-base mismatches and insertion-deletion loops arising from DNA replication. Heterozygous germline alteration of MMR genes (MSH2, MSH6, MLH1, PMS2) cause autosomal dominant Lynch syndrome (LS), most commonly manifesting as colonic or endometrial cancers, although brain, ovarian, and other organ systems may be involved. Neoplasia in LS usually arises after the age of 30 years. Constitutional mismatch repair deficiency (CMMRD) is inherited in an autosomal recessive manner due to biallelic germline alteration in one of the four MMR genes. Individuals with CMMRD typically develop cancer in the first decade of life, although some may present during the second decade. We present a series of five children who developed cancer prior to the age of 20 years (range: 2–12 years) with malignancies including colonic adenocarcinoma (N = 1), T-lymphoblastic lymphoma (N = 3), and high-grade glioma (N = 4). Two patients with MSH6 alterations developed a constellation of three primary tumors: high-grade glioma, T-lymphoblastic lymphoma, and colonic neoplasia including colonic adenocarcinoma in one patient and a tubular adenoma in the other.

Introduction

The DNA mismatch repair (MMR) system is responsible for correcting single-base mismatches and insertion-deletion loops that arise during DNA replication [1]. Germline alteration in genes encoding the MMR pathway (i.e. MSH2, MSH6, MLH1, PMS2, and rarely EPCAM) that result in defective MMR are associated with autosomal dominant Lynch Syndrome (LS) and autosomal recessive constitutional MMR deficiency (CMMRD) syndrome [2, 3]. LS is associated with high risk of malignancy, particularly involving the colon, endometrium, ovaries, brain, or other organ systems. CMMRD is characterized by a highly penetrant predisposition to a broad spectrum of early-onset malignancies, including hematological malignancies, brain tumors, colorectal and other gastrointestinal cancers. While tumors in LS usually arise in the third decade of life or later, CMMRD often manifests within the first or second decades [4]. Patients with CMMRD may also exhibit features overlapping with neurofibromatosis type 1 (NF1), including café-au-lait macules and Lisch nodules, which may complicate diagnosis [5]. CMMRD is an extremely rare disorder, with only about 300 pediatric and young adult cancer patients reported [6, 7]. This series describes the clinical, molecular, and familial features of five individuals diagnosed with CMMRD at two large children’s hospitals with findings that align with previous reports.

Materials and methods

Patients less than 20 years old carrying homozygous or compound heterozygous variants in MMR genes were identified at two large children’s hospitals via retrospective chart review from 2011 to 2021. Clinicopathologic features of each patient, including family history, demographic data, clinical features, tumor types, molecular variants, pathology reports, genetic test reports, and outcomes were extracted from medical records. The study was conducted in accordance with both institutional policies and applicable regulations.

Results

Five patients with CMMRD were identified, each demonstrating alterations either in MSH6 (homozygous: N = 3, compound heterozygous: N = 1) or PMS2 (homozygous, N = 1) (Table 1). The average age at presentation was 8 years (range: 2–12 years) and four of the five patients were male. Consanguinity was confirmed in four of the five cases. Café-au-lait spots were seen in all five patients, and two patients showed additional findings typically associated with neurofibromatosis type 1 (NF1) (e.g., Lisch nodules, optic glioma, axillary freckling). Patient 1 had a clinical diagnosis of NF1 in September 2019, genetic testing completed in January 2020, and the patient was diagnosed with CMMRD by genetics four months later (April 2020). Patient 2 had a clinical diagnosis of NF1 in 2008, genetic testing for NF1 (negative) in July 2009, and was diagnosed with CMMRD by genetics (Commercial Lynch Syndrome panel) eight years later (July 2017). Patients 3, 4 and 5 did not have formal genetic testing for NF1. Patient 4 with MSH6 alteration showed concomitant dextrocardia and primary ciliary dyskinesia. In contrast to colonic or endometrial cancers as the most common presenting tumors in LS, high-grade glial/glioneuronal neoplasms (n = 4) (Fig. 1) and lymphoblastic lymphomas (n = 3) (Fig. 2) were the most common tumors among these five patients with CMMRD. Three patients also developed gastrointestinal neoplasia, diagnosed as colonic adenocarcinoma in one patient with MSH6 alterations (Fig. 3), and tubular adenomas in two patients (one with MSH6 and one with PMS2 alterations). Three patients harboring MSH6 alterations are deceased (mean duration of survival after diagnosis of first malignancy: 3 years; range: 1 to 6 years), one with MSH6 alteration is alive with malignancy (1 year following diagnosis of first malignancy), and the patient with PMS2 alteration is alive with no evidence of malignancy (13 years following diagnosis of first malignancy).

Table 1.

Patient characteristics of the study cohort

Patient no.	Gender	Clinical Features	Diagnosis	Age at presentation (y)	Age at CMMRD dx	MMR Germline Variation	Family History	Status (age at last follow up [y])
1	F	Café-au-lait spots, Lisch nodules, and optic glioma (testing for germline NF1 variation was negative)	IDH-WT glioblastoma. T-LL Tubular adenomas	9 10 10	9	MSH6  (NM_000179.3):c.3939_3957dupTCAAAAGGGACATAGAAAA p.Ala1320fs  Homozygous	Consanguinity	D (10)
2	M	Innumerable café-au-lait spots (L > R), axillary/inguinal freckling, few areas of hypopigmentation (right back, lower extremities), no neurofibromas. Testing for germline NF1 variation mutation was negative.	Astrocytoma T-LL CRC	12 13 17	16	MSH6   (NM_000179.3):c.3768T>G p.Tyr1256Ter (NM_000179.3):c.3939_3957dup p.Ala1320SerfsTer Compound Heterozygous 1 variant confirmed maternal in origin; father not available for confirmation of other variant		D (18)
3	M	Café-au-lait spots	B-LL Glioneuronal (vs. embryonal CNS tumor)	5 7	5	MSH6 c.3680T>G, p.Ile1227Arg Homozygous	Consanguinity	D (7)
4	M	Café-au-lait spots Dextrocardia and primary ciliary dyskinesia; dextrocardia diagnosed at birth	IDH-mutant anaplastic astrocytoma (gliomatosis cerebri). Tumor mutations in IDH1, MSH6, ATRX, NF1, TP53	12	12	MSH6 c.3863_3865dupAAT, p.Phe1289 Homozygous	Consanguinity	D (13)
5	M	Café-au-lait spots	T-LL Tubular adenomas	2, relapse at 5y 6	6	PMS2 exon 14-15 deletion Homozygous	Consanguinity	ANEM (15)

Abbreviations: AWD Alive with disease, D Deceased, ANEM Alive with no evidence of malignancy

Fig. 1.

Glioblastoma demonstrating microvascular proliferation (a; arrowhead) and early tumor necrosis (a; star), marked nuclear anaplasia (b) and tumor multinucleated giant cells (b; arrow) and increased mitoses including abnormal mitosis (b; arrowhead) {Created in https://BioRender.com} [Hematoxylin and Eosin stain; a: 200x and b: 400x magnification]

Fig. 2.

Representative Hematoxylin and Eosin-stained microscopic image (a) demonstrating sheets of small sized lymphoma cells with admixed phagocytic macrophages giving a “Starry sky” appearance (400x magnification); Diff-Quik stained Cytospin preparation used (B; 1000x magnification) showing lymphoblasts and frequent mitoses (arrowheads). {Created in https://BioRender.com}

Fig. 3.

Invasive colonic adenocarcinoma with mucinous features characterized by infiltrating glands within a desmoplastic stroma (a; Hematoxylin and Eosin stain at 20x magnification); MSH6 immunohistochemistry demonstrating diffuse loss of nuclear expression within tissue (b; 100x magnification); positive MSH6 control included in figure b inset {Created in https://BioRender.com}

Discussion

This case series reinforces the variable phenotypic spectrum and early onset of malignancies in patients with CMMRD. All five patients developed tumors before age 13. Four patients developed lymphoblastic lymphomas, which aligns with prior studies showing that hematopoietic system malignancies are common and usually are non-Hodgkin lymphomas (NHL) [6]. According to Wimmer et al., NHL constitutes ∼5–7% of all childhood tumors but 14% of all CMMRD-associated malignancies, and most are derived from the T-cell lineage (i.e. T-cell lymphoblastic lymphoma) [6]. Four patients developed brain/central nervous system (CNS) malignancies, consistent with previous reports indicating that CNS tumors are common in CMMRD patients [6]. Wimmer et al. [6]. studied 149 patients with CMMRD and 81 of them showed brain/CNS tumors. One patient developed colorectal carcinoma (CRC), and two developed tubular adenomas, fitting into the designation of “LS-associated” neoplasia. In the Wimmer study, of the 149 CMMRD patients studied, 88 showed LS-associated carcinoma. All five of our patients showed skin manifestations (i.e. café-au-lait macules), which highlights the need to consider CMMRD in the differential diagnosis when NF1 is suspected but lacks a corresponding genetic diagnosis. Of note, congenital malformations have been described in a small proportion of CMMRD patients, most commonly including agenesis of the corpus callosum and gray matter heterotopia [6]. One patient in this case series had dextrocardia and primary ciliary dyskinesia. Three of our patients with MSH6 alterations are deceased, one with MSH6 alteration is alive with malignancy, and one with PMS2 alteration is alive with no evidence of malignancy. A recent comprehensive international study published by Ercan et al.. in 2024 investigated different outcomes of patients at age 15 with CMMRD based on their specific gene alteration. Their results indicated that individuals with PMS2 and MSH6 variations in CMMRD tend to have later cancer onset and better survival compared to those with MLH1 or MSH2 variations (overall survival of PMS2 alterations 63%, overall survival of MSH6 alterations 49%, overall survival of MLH1 alterations 19%, and overall survival of MSH2 0%) [7].

Diagnosis of CMMRD in a pediatric or young adult cancer patient has important implications for clinical management and surveillance. Familial genetic testing is necessary to determine inheritance status of MMR disease-associated variation upon identification in a proband. When each parent harbors a pathogenic MMR variant, each child has a 25% chance of being affected (inheriting both pathogenic alleles), a 50% chance of being an unaffected carrier (inheriting one pathogenic allele), and a 25% chance of inheriting neither allele. In this setting, siblings of a CMMRD patient have a 25% risk of having inherited the same variants and, hence, an equally high risk for childhood cancer [6]. Parents of CMMRD patients, who themselves harbor a single copy of an altered MMR gene, have an increased risk for LS-associated tumors in adulthood, as well as any additional heterozygous family members. Familial cascade testing of known disease-associated MMR variation is an important diagnostic test when coupled with genetic counseling to determine disease risk and the need for surveillance [8, 9]. The diagnosis of CMMRD can be challenging, particularly given the overlapping features with NF1 [5]. Therefore, diagnostic criteria for CMMRD have been suggested by the European consortium “care for CMMRD” [6]. The expert panel of the consortium established seven diagnostic criteria: four criteria with strong evidence of CMMRD (i.e., definitive diagnosis) and three criteria with moderate evidence (i.e., likely diagnosis), as outlined in Table 2. All criteria outlined warrant CMMRD surveillance [6]. They used three components to establish their criteria, namely (1) MMR germline results, (2) ancillary testing, and (3) clinical manifestations [6]. Multigene panel testing is recommended to investigate overlapping conditions which can mimic CMMRD [10]. They proposed a three-point scoring system to assess suspected CMMRD in pediatric or young adult cancer patients [6]. Tumors highly specific to CMMRD score 3 points, overrepresented malignancies score 2 points, and all others score 1 point [6]. Additional features—such as pigmentary skin changes, brain malformations, pilomatrixomas, a second childhood cancer, LS-associated tumors in relatives, or parental consanguinity—are assigned 1–2 points based on their CMMRD specificity and general population frequency [6]. A total score of 3 or more warrants suspicion of CMMRD [6].

Table 2.

Indication criteria for CMMRD testing in cancer patients

Indication for CMMRD testing in a cancer patient	≥ 3 points
Malignancies/premalignancies: one is mandatory; if more than one is present in the patient, add the points
Carcinoma from the LS spectrum* at age < 25 years	3 points
Multiple bowel adenomas at age < 25 years and absence of APC/MUTYH mutation(s) or a single high-grade dysplasia adenoma at age < 25 years	3 points
WHO grade III or IV glioma at age < 25 years	2 points
NHL of T-cell lineage or sPNET at age < 18 years	2 points
Any malignancy at age < 18 years	1 point
Additional features: optional; if more than one of the following is present, add the points
Clinical sign of NF1 and/or ≥ 2 hyperpigmented and/or hypopigmented skin alterations Ø>1 cm in the patient	2 points
Diagnosis of LS in a first-degree or second-degree relative	2 points
Carcinoma from LS spectrum* before the age of 60 in first-degree, second-degree, and third-degree relative	1 point
A sibling with carcinoma from the LS spectrum*, high-grade glioma, sPNET or NHL	2 points
A sibling with any type of childhood malignancy	1 point
Multiple pilomatricomas in the patient	2 points
One pilomatricoma in the patient	1 point
Agenesis of the corpus callosum or non-therapy-induced cavernoma in the patient	1 point
Consanguineous parents	1 point
Deficiency/reduced levels of IgG2/4 and/or IgA	1 point

Table used with permission from BMJ. Wimmer K, Kratz CP, Vasen HFA, et al. Journal of Medical Genetics 2014; 51:355–365

Abbreviations: LS Lynch syndrome, CMMRD Constitutional mismatch repair deficiency, NHL Non-Hodgkin’s lymphomas, sPNET Supratentorial primitive neuroectodermal tumors, LS spectrum Colorectal, endometrial, small bowel, ureter, renal pelvis, biliary tract, stomach, bladder carcinoma

Both immunohistochemistry (IHC) for MMR proteins on tissue sections and microsatellite instability analysis are diagnostic methods to substantiate the suspected diagnosis [6]. IHC is considered the preferred method since it may also guide targeted sequencing analysis and has been shown to render reliable results in most solid tumors [6]. In contrast to LS where expression loss is observed only in neoplastic cells, in CMMRD patients IHC detects expression loss of one (or two) of the MMR proteins in both neoplastic and non-neoplastic tissues [6]. Hence, negative IHC staining in neoplastic and surrounding normal cells should not be interpreted as a failure of proper staining and care should be taken to use an on-slide staining control from a different individual [6]. The final confirmation of CMMRD should come from the determination of the causative biallelic alterations in the patient [6].

Surveillance in CMMRD involves multiple organ systems and numerous recommendations have been published that include initiating screening for lymphomas and abdominal tumors with ultrasound at age 1, brain tumors with head ultrasound at birth and brain MRI at age 6 months, gastrointestinal tumors with endoscopy/colonoscopy and video capsule study at age 6 years, whole body MRI every 12 months starting at 6 years, and genitourinary tumors at age 12 years, as well as skin exam annually starting at diagnosis [11–14]. The youngest patient in our series presented with leukemia at 2 years old. In comparison, childhood LS-associated colorectal cancers remain rare so trends are not well defined [15]. Current guidelines call for screening colonoscopy to begin in patients with Lynch syndrome between age 20–25, or older based on specific genetic change, with repeat colonoscopies to occur every two years unless findings dictate otherwise [15–18]. In families with particularly early-onset colorectal cancers, screening should begin two to five years prior to the earliest colorectal cancer diagnosed before age 25 [15–18]. Genetic testing is typically offered after age 18 in families with known Lynch syndrome, to ensure appropriate autonomy and consent for testing; rarely, genetic testing will be offered at a younger age in specific circumstances [15, 16].

Conclusion

The wide variability in cancer types underscores the importance of early genetic testing in pediatric patients with multiple tumors or family histories suggestive of LS or CMMRD. Given the high risk of multiple primary malignancies, lifelong surveillance strategies and family genetic counseling are critical. Increased awareness and identification of clinical red flags may allow earlier diagnosis and potentially life-saving surveillance or prophylactic measures. The presence of NF1-associated clinical features among these patients with CMMRD is diverse and can involve multiple organ systems. Importantly, the presence of NF1-like features with negative genetic testing for NF1 variants should raise suspicion for other diseases such as CMMRD. Although these observations suggest an extraordinarily high tumor risk in CMMRD patients, data from a larger number of pre-symptomatically tested patients followed over a longer period of time are needed to assess the penetrance of CMMRD with respect to cancer development [6].

Authors’ contributions

J.V.G authored the first draft of the manuscript and created tables, S.J.L performed data collection and initial analysis A.S designed the study, supervised data analysis, revised the first draft and prepared figures 1 and 3.S.K prepared figure 2. All authors reviewed and edited the manuscript.

Funding

Molecular analyses for Patient 1 were supported by the Nationwide Foundation Pediatric Innovation Fund.

Data availability

DNA exome sequencing data for Patient 1 has been deposited to dbGaP under accession number phs001820.v1.p1.

Declarations

Ethics approval and consent to participate

This study was conducted under a research protocol approved by the Institutional Review Board at Nationwide Children’s Hospital (IRB17-00206).

Competing interests

The authors declare no competing interests.