Germline Alterations in RASAL1 in Cowden Syndrome Patients Presenting with Follicular Thyroid Cancer and in Individuals with Apparently Sporadic Epithelial Thyroid Cancer

Abstract

Context:

RASAL1 has recently been identified as an important tumor suppressor for sporadic thyroid tumorigenesis, particularly for follicular thyroid cancer (FTC) and anaplastic thyroid cancer. Thyroid cancer is an important component of Cowden syndrome (CS). Patients with germline PTEN mutations have an overrepresentation of FTC over other histological subtypes.

Objective:

To determine the prevalence of germline RASAL1 mutations in PTEN mutation–positive and wild type CS patients.

Setting and Design:

We reviewed our prospective database of more than 3000 CS/CS-like patients and retrospectively identified a subset of patients who presented with thyroid cancer for RASAL1 mutation analysis. We reviewed data from The Cancer Genome Atlas (TCGA) sporadic papillary thyroid cancer (PTC) database with germline data for RASAL1 mutations to determine the prevalence of germline RASAL1 mutations in CS-related thyroid cancer patients.

Results:

We scanned 155 CS/CS-like patients with thyroid cancer for germline RASAL1 mutations. Of the 155 patients, 39 had known germline pathogenic PTEN mutations (PTEN^mut+) and 116 were PTEN mutation negative (PTEN^WT). Among these 155 patients, we identified RASAL1 germline alterations suspected as being deleterious in two patients. Both were patients with PTEN^WT who had FTC (2/48, 4.1%). This was in contrast to patients with PTEN^mut+ who had thyroid cancer (0/39). Of 339 sporadic patients with PTC from the TCGA study, 62 (18%) had germline RASAL1 variants predicted to be deleterious. TCGA patients with follicular-variant PTC were statistically overrepresented (21/62, 34%) among patients with deleterious RASAL1 variants compared with those without (57/277, 21%).

Conclusions:

Germline RASAL1 alterations are uncommon in patients with CS but may not be infrequent in patients with apparently sporadic follicular-variant PTC.

The Ras pathway is one of the most commonly deregulated pathways in human cancer (1). The transforming function of oncogenic Ras has been initially attributed to its capacity to endow cells with sufficiency in growth signals. There is now considerable and increasing evidence that beyond Ras, the aberrant function of an expanding list of RAS superfamily proteins has been implicated in human cancer growth and development. However, whereas mutational activation of RAS is seen commonly in human cancers, direct mutation of other RAS superfamily GTPases is not seen frequently. Recently, Xing and colleagues (2) studied how alternative RAS signaling–related genes affected thyroid tumorigenesis. Compared with normal human thyroid tissue, the RAS GTPase–activating protein (RasGAP) gene, RASAL1, was seen to be commonly but differentially somatically mutated or hypermethylated in sporadic thyroid cancers. Sequence analysis of primary thyroid cancer samples for somatic mutations in RASAL1 identified mutations predominantly in anaplastic thyroid cancer (ATC) and follicular thyroid cancer (FTC). All the somatic mutations were located in the RAS GTPase–activating domain of RASAL1, with majority of the missense mutations located at highly conserved sites, and not surprisingly, functional characterization of RASAL1 mutants demonstrated their impaired ability to suppress both RAS-coupled MAPK and PI3K pathways.

Ras proteins cycle between “on” and “off” conformations that are conferred by the binding of GTP and GDP, respectively. Loss of RasGAP activity allows uncontrolled GTPase activity and can promote tumorigenesis leading to tumor formation. As such, RasGAPs are well placed to function as potential tumor suppressors. Indeed, NF1 encodes a RasGAP and is mutated in neurofibromatosis type 1 (3), a known cancer predisposition syndrome with increased risk of breast cancer and glioblastomas. It is not known whether germline mutations in RASAL1 will similarly predispose to cancers.

Epithelial thyroid carcinomas are a major component of Cowden (CS) and related syndromes characterized by germline mutations in the PTEN tumor suppressor gene (4, 5). CS is an autosomal-dominant disorder characterized by the development of multiple hamartomas, and importantly, carcinomas of the thyroid, breast, endometrium, and kidney (5, 6). Germline mutations in PTEN increase the risk of epithelial thyroid cancer by more than 70-fold when compared with that of the general population with an overrepresentation of FTC over other histology subtypes (4). Among patients who present with CS phenotypes, 25% have germline mutations in PTEN (7). More recently, germline mutations in PI3KCA and AKT1 have also been implicated in CS (8) and it is likely that in time, other candidate genes in both canonical PTEN/PI3KCA/AKT pathways and beyond may be found to be associated with CS. RASAL1 alterations may preferentially result in activation of the PI3K pathway over the MAPK pathway as evidenced by the overrepresentation of RASAL1 alterations seen in FTC and ATC over papillary thyroid cancer (PTC). This aligns well with the fact that RAS itself is a classic dual activator of MAPK and PI3K-AKT pathways but in thyroid cancer, it seems that RAS mutations seem to preferentially activate the PI3K-AKT pathway (9, 10). Given these recent findings, we explored the prevalence of germline RASAL1 mutations in a subset of patients with CS who had thyroid cancer.

Materials and Methods

Patients

Probands who met at least the relaxed International Cowden Consortium operational criteria for CS were eligible. Relaxed criteria are defined as full criteria minus one criterion and such individuals are referred to as CS-like. For each patient, the medical record was examined by cancer genetics professionals and when possible, primary documentation of medical records/pathology reports were obtained for confirmation of the thyroid cancer and precise histology, with the patients' consent. For all patients, a semiquantitative score, the Cleveland Clinic score based on clinical features, was calculated. The Cleveland Clinic score has been shown to provide a well-calibrated estimation of pretest probability of PTEN status (7), the higher the score the higher the likelihood for detecting a germline PTEN mutation. We reviewed our database of more than 3000 CS/CS-like patients and identified a subset of patients who presented with thyroid cancer for RASAL1 mutation analysis. For patients with pathogenic germline PTEN mutations (PTEN^mut+), we included all patients who had a history of thyroid cancer; for patients who were PTEN mutation negative (PTEN^WT), we selectively enriched for histologies which were found to harbor somatic RASAL1 mutations. We included all PTEN^WT patients who had FTC (n = 48), ATC (1) histologies from our database for analysis and complemented that with a subset of those with PTC and other histological subtypes for exploration.

PTEN mutation/deletion and RASAL1 mutation analysis

All research participants underwent prior PTEN (NM_ 000314.4) mutation analysis, described as follows. Genomic DNA was extracted from peripheral blood leukocytes using standard methods (11). Scanning of genomic DNA samples for PTEN mutations was performed as previously reported with a combination of denaturing gradient gel electrophoresis, high-resolution melting curve analysis (Idaho Technology), and direct Sanger sequencing (ABI 3730xl, Life Techonologies). Deletion analysis using the multiplex ligation-dependent probe amplification (MLPA) assay (12) was performed with the P158 MLPA kit (MRC-Holland) according to manufacturer's protocol. All patients underwent PCR-based Sanger sequencing of the PTEN promoter region as previously described (13). Eligible patients had subsequent RASAL1 (NM_004658.2) mutation analysis. Exons 2–22 that span the whole coding region of RASAL1 were PCR amplified and Sanger sequenced. Nonsynonymous, frameshift, splice-site, nonsense mutations as well as large deletions and whole gene-deletion are assigned as mutation positive for both PTEN and RASAL1. All intronic and synonymous mutations were classified as variants of unknown significance and considered as mutation negative for both PTEN and RASAL1. Prediction databases were used to assist missense mutation annotations for both genes (14–16). To be conservative, all cases without proof of functionality or that were predicted to be nonpathogenic in two of three prediction databases were considered as variants of unknown significance and, for the purposes, of this study were considered as mutation negative. Fisher's exact test was used to measure the association between germline RASAL1 mutation status and histological subtypes.

Results

We scanned 155 CS/CS-like patients with thyroid cancer for germline RASAL1 mutations. Of the 155 patients, 39 were PTEN^mut+ and 116 were PTEN^WT. The pathology and clinical characteristics are shown in Table 1. Of the 116 PTEN^WT CS/CS-like patients who presented with thyroid cancer, 53 (46%) had either FTC or follicular-variant PTC (FvPTC), 54 (47%) had PTC, 7 (6%) had Hurthle cell thyroid cancer, and 1 had ATC. Of the PTEN^mut+ patients with thyroid cancer, a third (13/39) had FTC or FvPTC (Table 2).

Table 1.

Prevalence of Germline RASAL1 Alterations in CS Patients Screened (n = 195)

Cohort	Thyroid Cancer Histology	Female	Male	RASAL1 Mutations
PTEN mutation negative (n = 116)	FTC	48	1	c.982C>T; R328W (2)
	FvPTC		4
	Anaplastic	1
	Hurthle	7
	PTC	38	16
PTEN mutation positive (n = 39)	FTC	2	2
	FVPTC	6	3
	Anaplastic		1
	PTC	19	4
	NOS	2
“Control” elderly no-cancer PTEN wild typea (n = 40)	Nil	35	5

NOS, not otherwise specified; CC, Cleveland Clinic PTEN Risk Score.

^aAge at CS diagnosis ≥ 60 y and CC ≥ 10.

Table 2.

Clinical Characteristics of Patients with Germline RASAL1 Alterations

CCF ID	CC Score	Sex	Clinical Features	RASAL1	Protein Prediction
					SIFT	PolyPhen 2	Condel
6098	15	F	FTC (age 41 y); goiter; macrocephaly; fibroids; hemangioma	c.982C>T; R328W	Deletrious	Deleterious	Deleterious
7119	4	F	FTC (age 59 y); breast cancer (lobular + contralateral infiltrating ductal carcinoma at age 62 y); fibroids; benign breast disease	c.982C>T; R328W	Deleterious	Deleterious	Deleterious

Of the 155 patients, we identified RASAL1 germline alterations suspected as being deleterious in two patients (Table 3). Interestingly, both were PTEN^WT patients who had FTC (2/48; 4.1%) and had the same missense RASAL1 change (R328W) in exon 11 within the RASGap superfamily domain. This alteration was predicted to be pathogenic across three protein prediction databases (Table 2). We did not see any deleterious germline RASAL1 alterations in PTEN^mut+ patients with thyroid cancer.

Table 3.

Clinical Characteristics of Patients with TCGA PTC with Germline RASAL1 Analysis (n = 339)

Clinical Characteristics	All Patients (%) (n = 339)	Deleterious RASAL1 Variants (%) (n = 62)	No Deleterious RASAL1 Variants (%) (n = 277)	P Value
Age
<46 y	173 (51.0)	35 (56.4)	138 (49.8)	.500
≥46 y	166 (48.9)	27 (43.5)	139 (50.1)
Gender
Male	88 (26.0)	19 (30.6)	69 (24.9)	.569
Female	251 (74.0)	43 (69.3)	208 (75.1)
Ethnicity
White	218 (62.3)	37 (59.6)	181 (65.3)	.471
Asian	31 (9.1)	4 (6.5)	27 (9.7)
African American	11 (3.2)	1 (1.6)	10 (3.6)
Unknown	84 (24.8)	20 (32.2)	64 (23.1)
Histological subtype
Follicular Variant	78 (23.0)	21 (33.9)	57 (20.6)	.027
Classical	228 (67.3)	40 (64.5)	188 (67.9)
Tall Cell Variant	27 (8.0)	1 (1.6)	26 (9.4)
Other	6 (1.8)	0	0
First-degree family with thyroid cancer
Yes	14 (4.1)	2 (3.2)	12 (4.3)	1.000
No	325 (95.9)	60 (96.8)	265 (95.6)

Fisher's exact test was used to assess the significance of deleterious RASAL1 variants among patients for each clinical variable. Figures in bold indicate a P value of <.05

We accessed The Cancer Genome Atlas (TCGA) PTC database for those who had DNA from blood (representing the germline) analyzed to determine whether germline RASAL1 alterations were seen in the TCGA cohort. Of the 339 TCGA PTC patients who had blood analyzed, 62 (18%) had germline RASAL1 variants predicted to be deleterious, with two of these 62 having the same germline R328W alteration found in our CS-FTC patients. We then examined whether any demographic or clinical factor (Table 3) was predictive of presence of deleterious germline RASAL1 alterations. Age, sex, ethnicity, and family history of thyroid cancer in a first-degree relative (Table 3) was not predictive. Only FvPTC histology was statistically overrepresented among patients with deleterious germline RASAL1 alterations compared with those without FvPTC histology (34 vs. 21%, P = .027).

Discussion

Given the increasing realization of the role of Ras regulatory proteins and their role in cancer, and the down-regulation of RASAL1 expression in a number of cancers (2, 17), we sought to determine what role if any germline RASAL1 germline mutations may play in CS-related thyroid cancer. CS is the ideal hereditary cancer syndrome to explore whether RASAL1 could be responsible for a subset of PTEN^WT CS patients. Given the recent convincing findings of RASAL1's somatic role as a major tumor suppressor in sporadic thyroid cancer by affecting both RAF/MEK/ERK and PI3K/AKT/MTOR pathways, particularly FTC and ATC (2), we wanted to explore whether germline RASAL1 mutations is implicated specifically in CS-related thyroid cancer. Although our results yielded only two patients with deleterious germline RASAL1 alterations, it is intriguing that both patients had FTC over other histological subtypes. R328W is rarely seen in the general population with a minor allele frequency of less than 0.001 from the 1000 Genomes Project (18) so it is in our view significant that we saw it in 4% PTEN^WT patients who presented with FTC and none in the other histological subtypes. Unfortunately, although FTC is overrepresented in patients with known pathogenic PTEN mutations (4), it is still the less commonly encountered histological subtype in our PTEN^WTcohort and we are unable to expand the sample size for RASAL1 screening. Although our numbers are relatively small and at this juncture, any correlation between germline RASAL1 alteration and FTC speculative, it is encouraging that we did not see any deleterious alterations in the other patients screened. We had scanned our PTEN^mut+ patients.

Coexisting germline mutations are not commonly encountered. We and others have reported on several such occurrences (19–22). Potentially, we will see more cases with the increased use of multigene panels and exome/genome sequencing highlighting the need to better understand the relative contributions these alterations play in tumorigenesis. Our recent work reveals that a subset of CS patients with germline variants in succinate dehydrogenase genes (SDHx) had loss of steady-state p53. Interestingly, our published data suggest that the potential regulation of HIF1α, p53 and PTEN signaling by mitochondrial metabolism may play a role in CS tumorigenesis (23). The prevalence of thyroid cancer in patients with both germline PTEN mutations and SDHx variants trends higher than that of PTEN^mut+-only patients, and the histology of these six PTEN/SDHx-related thyroid cancers was exclusively papillary. We had chosen to include the PTEN^mut+ cohort in our present study because we wanted to explore whether RASAL1 alterations modified the risk preferentially for FTC or ATC but did not find any such associations.

Because we were concerned that our numbers are small, we analyzed an additional 40 CS and CS-like patients who were 60 years of age or older with no cancers as a further control group and again we saw no alterations in this group of patients. We acknowledge that our study would be strengthened by increased sample size and the availability of tissue samples for further validation. To overcome this, we accessed the TCGA database for those who had blood DNA analyzed. Interestingly, the specific germline R328W alteration is seen in 4% (2/48) of our PTEN^WT FTC patients compared with 0.6% (2/339) TCGA-sporadic PTC patients (P = .077). Importantly, among these 339 sporadic PTC patients from the TCGA study, 62 (18%) had germline RASAL1 variants predicted to be deleterious. Strikingly, TCGA patients with FvPTC were statistically significantly overrepresented (21/62, 34%) among patients with deleterious RASAL1 variants compared with those without (57/277, 21%). This observation is interesting given the clear tumor suppressor role RASAL1 plays in FTC. It would be interesting when the TCGA project expands to include FTC to see whether germline RASAL1 alterations would be more prevalent aligned with our hypothesis. We can conclude that germline RASAL1 alterations are uncommon in CS patients but may be relatively frequent in patients with apparently sporadic thyroid carcinomas with follicular features, here, FvPTC. Therefore, future studies exploring the germline role of RASAL1 may wish to selectively evaluate CS-related FTC patients who are PTEN mutation negative and apparently sporadic follicular types of thyroid cancer, both FTC and FvPTC compared with conventional PTC.

Funding Statement

This work was supported by NCI grant P01CA124570 (C.E.). J.N. is the Ambrose Monell Foundation Cancer Genomic Medicine Clinical Fellow at the Cleveland Clinic Genomic Medicine Institute. Y.N. is a CoGEC Scholar funded, in part by, NCI grant R25TCA094186. C.E is the Sondra J. and Stephen R. Hardis Chair of Cancer Genomic Medicine at the Cleveland Clinic and is an American Cancer Society Clinical Research Professor, generously funded in part, by the F.M. Kirby Foundation.