http://orcid.org/0000-0003-0105-9760
Figures
Abstract
Cyclin D1 (CCND1) is a core cell cycle regulator and is frequently overexpressed in human cancers, often via amplification, translocation or post-transcription regulation. Accumulating evidence suggests that mutations of the CCND1 gene that result in nuclear retention and constitutive activation of CDK4/6 kinases are oncogenic drivers in cancer. However, the spectrum of CCND1 mutations across human cancers has not been systematically investigated. Here, we retrospectively mined whole-exome sequencing data from 124 published studies representing up to 29,432 cases from diverse cancer types and sites of origin, including carcinoma, melanoma, sarcoma and lymphoma/leukemia, via online tools to determine the frequency and spectrum of CCND1 mutations in human cancers and their associated clinico-pathological characteristics. Overall, in contrast to gene amplification, which occurred at a frequency of 4.8% (1,419 of 28,769 cases), CCND1 mutations were of very low frequency (0.5%, 151 of 29,432 cases) across all cancers, but were predominantly enriched in uterine endometrioid-type adenocarcinoma (6.5%, 30 of 458 cases) in both primary tumors and in advanced, metastatic endometrial cancer samples. CCND1 mutations in endometrial endometrioid adenocarcinoma occurred most commonly in the c-terminus of cyclin D1, as putative driver mutations, in a region thought to result in oncogenic activation of cyclin D1 via inhibition of Thr-286 phosphorylation and nuclear export, thereby resulting in nuclear retention and protein overexpression. Our findings implicate oncogenic c-terminal mutations of CCND1 in the pathogenesis of a subset of human cancers and provide a key resource to guide future preclinical and clinical investigations.
Citation: Xu J, Lin DI (2018) Oncogenic c-terminal cyclin D1 (CCND1) mutations are enriched in endometrioid endometrial adenocarcinomas. PLoS ONE 13(7): e0199688. https://doi.org/10.1371/journal.pone.0199688
Editor: Aamir Ahmad, University of South Alabama Mitchell Cancer Institute, UNITED STATES
Received: March 2, 2018; Accepted: June 12, 2018; Published: July 3, 2018
Copyright: © 2018 Xu, Lin. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. Somatic CCND1 mutation data and clinico-pathological characteristics for the 29,432 patients analyzed from the original 124 publications in this study are also freely available and can be found on the cBioPortal via the following specific link: http://bit.ly/2BjBlBs. Similarly, CCND1 gene amplification data for the 28,769 patients analyzed from the original 124 publications are also freely available and can be found via the following cBioPortal link: http://bit.ly/2AJ1eJW.
Funding: The authors received no specific funding for this work.
Competing interests: Author DIL recently became a molecular tumor board consultant for Foundation Medicine, Inc, in Cambridge, MA. However, DIL has not received any funding or payments from Foundation Medicine for this study. Furthermore, this does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Cyclin D1, encoded by the CCND1 gene, is a core cell cycle regulator that promotes cellular proliferation and is frequently overexpressed in human cancers. Alterations of cyclin D1 resulting in nuclear protein expression is thought to promote tumorigenesis in various types of cancers[1][2], often as a result of CCND1 gene amplification, such as in breast, lung, bladder carcinomas [3][4], or as a result of CCND1 translocation in mantle cell lymphoma [5]. Accumulating evidence also suggests that mutations of the CCND1 gene that result in nuclear retention and constitutive activation of CDK4/6 kinases are oncogenic drivers in cancer [6][7][8][9]. Oncogenic mutations of CCND1 have been previously reported in endometrial and esophageal carcinomas [10][11][12], however, the spectrum of CCND1 mutations across human cancers has not been systematically investigated.
The c-terminus of cyclin D1 contains a nuclear export signal and a phospho-degron, which is a region involved in cyclin D1 phosphorylation, nuclear export, ubiquitination and protein degradation. During the S phase of the cell cycle, phosphorylation of cyclin D1 at Thr-286 by GSK3β triggers cyclin D1 nuclear export and ubitiquin-mediated proteolysis via the 26S proteasome in the cytoplasm [13][14][15]. For this reason, mutations in the c-terminus of cyclin D1 may result in oncogenic activation of cyclin D1 via inhibition of Thr-286 phosphorylation, nuclear export and protein degradation, thereby promoting cyclin D1 nuclear accumulation [6]. Nuclear retention of cyclin D1 constitutively activates CDK4/6 kinases through the cell cycle and thereby promotes cellular proliferation and tumorigenesis [12].
Specific CCND1 c-terminal mutations involving Thr286 and Pro287 have been previously reported in endometrial cancer, such as T286I, P287T, P287S, which resulted in nuclear accumulation of cyclin D1, gain-of-function and cellular transformation [10][11]. Endometrial carcinoma is the most frequently diagnosed type of gynecological malignancy, with ~40,000 new cases and 7,500 mortalities reported in the United States in 2008 [16]. Currently, there are no FDA-approved or NCCN-compendium listed treatments specifically for patients with CCND1-mutant uterine endometrioid carcinoma. However, there are clinical trials evaluating palbociclib, a CDK4/6 specific inhibitor (codenamed PD-0332991, trade name Ibrance) [2][17], including a phase 2 trial in patients with metastatic endometrial cancer (ClinicalTrials.gov Identifier: NCT02730429), underscoring the importance of identifying mechanisms of CDK4 activation in endometrial cancer.
Recently, The Cancer Genome Atlas (TCGA) project and other whole exome studies, such as MSK-IMPACT trial, have molecularly characterized many different types of tumors, including endometrial adenocarcinomas [18][19]. The TCGA Data Portal and other publicly available online data mining programs offer a host of tools that can be used to interrogate specific genetic events relative to survival and clinico-pathological characteristics, as well as to generate hypotheses for future testing on a variety of tumor types [20][21].
The goal of this study was to mine whole-exome sequencing data from published studies representing a large number of cases and a plethora of cancer diagnoses to determine the frequency and spectrum of activating CCND1 mutations across human cancers. Here we demonstrate that activating, putative driver CCND1 mutations were of low frequency across all cancer types, but were predominantly enriched in uterine endometrioid-type adenocarcinoma.
Methods
Study type and design
With approval of a BIDMC/DFCI Institutional Review Board, we performed an observational retrospective secondary data analysis of clinically-annotated multi-platform cancer–omics datasets from the Cancer Genome Atlas Project (TCGA) and other previously published cancer genomics datasets. A total of 124 published studies with sequencing data (S1 Table), representing 29,432 cases from diverse cancer types and sites of origin, such as carcinoma, melanoma, mesothelioma, germ cell tumor, sarcoma and lymphoma/leukemia, were interrogated for CCND1 mutations. We exploited public available online tools and resources in order to data mine the cancer datasets with respect to CCND1 alterations, to interrogate different clinic-pathological characteristics relative to genomic alterations of CCND1, and to confirm the effect of specific CCND1 mutations on its biological and oncogenic function.
Cancer cohorts, gene amplification and gene mutation
Clinical and molecular data from up to 124 previously published cancer genomics studies (S1 Table) was analyzed with respect to CCND1 amplification and mutations via the cBio Cancer Genomics Portal (http://cbioportal.org; Memorial Sloan Kettering Cancer Center, New York, NY) [20][21]. The cohorts were queried for CCND1 gene amplification or mutations in the cbio portal by using the advanced onco query language (CCND1: AMP or CCND1: MUT) [20][21]. In this manuscript, copy number gain was defined as gain of one gene copy number, while amplification was defined as gain of 2 or more gene copy numbers. Online analysis was performed between January 1st, 2015 and May 7th, 2018.
CCND1 mutations specifically in primary endometrial carcinoma were interrogated via the cbio portal and via the advanced onco query language specifically in the endometrial carcinoma TCGA study with sequencing data (n = 248) [18]. In a similar manner, CCND1 mutations in advanced, metastatic endometrial carcinoma samples were interrogated specifically in the MSK IMPACT clinical cohort of advanced/metastatic endometrial carcinoma cases (n = 210) via the cBio portal [19].
In the primary endometrial carcinoma TGCA cohort, the following molecular and histological subsets were also analyzed via the cBio portal and by using the advanced onco query language: 1) complete endometrioid-type tumors (n = 232), 2) copy number high tumors (n = 60), 3) copy number low tumors (n = 90), 4) MSI (hypermutated) tumors (n = 65), 5) POLE (ultramutated) tumors (n = 17), and 6) tumors with specific serous-type histology (n = 53).
Review of pathological data
Pathology reports from the endometrial carcinoma TCGA corhort were downloaded via the cBio portal. Quality control hematoxylin and eosin (H&E) images from frozen and permanent sections from TCGA cohort were re-analyzed by a board-certified gynecologic pathologist (D.L.) to confirm histological subtype via the TCGA Biosig website hosted at Lawrence Berkeley National (http://tcga.lbl.gov:8080/biosig/tcgadownload.do), and via the Cancer Digital Slide Archive (http://cancer.digitalslidearchive.net; Emory University, Atlanta, GA). Tumor stage and overall FIGO grade were also independently re-evaluated whenever possible; however, they were deferred to the available pathology reports since only limited virtual images from each case were publicly available for review. For the MSK IMPACT cohort of advanced, metastatic endometrial cancer, histological subtype was deferred to the reported subtype since H&E images were not publicly available for re-review.
Analysis of CCND1 mutations
Somatic CCND1 mutations derived from exome studies, including from TCGA and MSK IMPACT endometrial carcinoma cohorts, were retrieved via the cBio portal (http://www.cbioportal.org/public-portal/). The effect of specific somatic mutations on CCND1 function was determined by cross-referencing them to previously published reports, wherein the effect of specific CCND1 mutations on Cyclin D1 activation or gain-of-function was demonstrated either in vitro or in vivo [6][12][11]. In addition, effects of CCND1 mutations were independently queried in the online Precision Oncology Knowledge Base tool (http://www.OncoKB.org), which is a publicly available, evidence-based web resource tool that annotates and curates the biologic and oncogenic effects, as well as the predictive significance of somatic molecular alterations present in patient tumors with the goals of facilitating the interpretation of genomic alterations and supporting optimal treatment decisions [22].
Results
Frequency of CCND1 mutations across cancer
To assess frequency and spectrum of CCND1 mutations across cancer, molecular data from 124 studies (S1 Table), representing 29,432 cases/patients from diverse cancer types and sites of origin, were mined via the cBio Cancer Genomics Portal (http://cbioportal.org) [20][21]. Somatic mutation of the CCND1 gene was seen in 0.5% (151 of 29,432) of cases spanning all regions of the cyclin D1 protein (Table 1 and Fig 1A). By combining the molecular data of studies from the same cancer subtype and site of origin, the most frequently CCND1-mutated cancer subtypes were mantle cell lymphoma (34.5%, 10 of 29 cases), endometrial adenocacinoma (6.5%, 30 of 458 cases), multiple-myeloma (3.3%, 7 of 205 cases), colorectal carcinoma (1.1%, 23 of 2067 cases), melanoma (1.1%, 8 of 708 cases), skin non-melanoma cancer (1.1%, 2 of 177 cases) and uterine sarcoma (1.1%, 1 of 93 cases).
Across all human cancers analyzed (A), or specifically in the mantle cell lymphoma cohort (B), in the primary endometrial endometrioid adenocarcinoma TCGA cohort (C), and in the advanced or metastatic endometrial carcinoma MSK-IMPACT cohort (D).
The most commonly mutated amino acid of cyclin D1 was Proline 287, followed by Threonine 286, both of which are located in the c-terminus of cyclin D1 (Fig 1A). The third most common mutated amino acid was Tyrosine 44, which is located in the n-terminus of cyclin D1 (Fig 1A). The location of CCND1 mutations appeared to be tissue specific. In mantle cell lymphoma and multiple myeloma, the majority of CCND1 mutations occurred within the amino terminal domain of cyclin D1 (S2 Table and Fig 1B), while in endometrial adenocarcinomas, and less frequently in colorectal carcinoma and melanoma, CCND1 mutations centered on the carboxy terminal domain (Fig 1C and 1D and S1 Fig). These findings support the notion that endometrial adenocarcinomas are enriched for a unique spectrum of somatic CCND1 alterations, in which a significant number of CCND1 mutations occur in the c-terminal domain of cyclin D1 compared to other cancer subtypes.
Frequency of CCND1 amplification in cancer
Somatic CCND1 mutations occurred less frequently than CCND1 gene amplification. Overall, in contrast to somatic mutations, CCND1 gene amplification occurred at a frequency of 4.8% (1,419 of 28,769 cases/patients) across all cancers and studies with gene amplification data (Table 2). By combining the molecular data of studies from the same cancer subtype and site of origin, the most frequently CCND1-amplified cancer subtypes were head and neck squamous cell carcinoma (27.6%, 77 of 279 cases), breast cancer (16.7%, 735 of 4406 cases), bladder cancer (6.9%, 65 of 938 cases), esophageal-gastric cancer (4.8%, 72 of 1494 cases), non-small cell lung cancer (4.6%, 168 of 3648 cases), melanoma (3.3%, 23 of 708 cases), ovarian cancer (2.5%, 14 of 551 cases), appendiceal cancer (2.5%, 2 of 79 cases), hepatobiliary cancer (2.4%, 24 of 997 cases), prostate cancer (2.4%, 54 of 2262 cases) and endometrial cancer (2.3%, 11 of 480 cases).
CCND1 mutations in primary endometrial adenocarcinomas
Our analysis of CCND1 mutations subsequently focused on two endometrial adenocarcinoma cohorts, since this cancer subtype was most frequently enriched for c-terminal cyclin D1 mutations (Fig 1C and 1D). First, we analyzed the primary endometrial adenocarcinoma (EMCA) cohort (n = 248) of the Cancer Genome Atlas Project (TCGA) study [18], in which CCND1 was mutated in 6.1% (15 of 248) cases (Table 1). In the TCGA EMCA cohort, the majority of CCND1 mutations (60%, 9 of 15) were located in the carboxy terminal region of cyclin D1 (Fig 1C and Table 3). Next, the pathological data was re-reviewed in terms of patient age, tumor type, stage and FIGO grade. Patient age in CCND1-mutated cases ranged from 33 to 76 years (Mean = 56; median = 58) (Table 3). The majority of CCND1-mutated cases were of stage I (73%, 11 of 15) and FIGO grade 1 or 2 tumors (40% and 47% respectively) (Table 3). A smaller percentage of CCND1-mutated primary TGCA EMCA tumors were stage III (20%, 3 of 15), stage IV (7%, 1 of 15) or FIGO grade 3 (13%, 2 of 15). For all CCND1-mutated TGCA EMCA cases, the tumors were of endometrioid histological subtype, which were confirmed by reviewing available H&E images (Fig 2).
A (40x, case D1-A0ZO), B (400x, case D1-A0ZO) and C (40x, case B5-A0JV): grade 1 endometrioid adenocarcinoma containing T286I mutations. D (40x, case D1-A0ZU): grade 1 endometrioid adenocarcinoma containing P287S mutation. E (40x, case D1-A17D) and F (400x, case D1-A17D): grade 3 endometrioid adenocarcinoma containing P287S mutation.
The TCGA has previously stratified endometrioid endometrial adenocarcinomas into 4 different molecular subgroups: 1) low copy-number alterations, 2) high copy-number alterations (serous-like), 3) POLE ultramutated and 4) microsatellite instability hypermutated [18]. Within this molecular stratification system, we identified c-terminal cyclin D1 mutations in the low copy-number alterations and in the microsatellite instability hypermutated molecular subgroups (data not shown), both of which have intermediate prognosis. In contrast, no c-terminal cyclin D1 mutations were identified in the POLE ultramutated (best prognosis) and in the copy-number high, serous-like (worst prognosis) molecular subgroups (data not shown) or in TCGA tumors with specific serous-type histology (data not shown).
CCND1 mutations in advanced or metastatic endometrial carcinoma
We next interrogated the endometrial carcinoma cohort of cases from the MSK-IMPACT study in order to determine whether CCND1 c-terminal mutations also occurred in the advanced or metastatic endometrial carcinoma setting [19]. In this cohort of advanced EMCA (n = 210), CCND1 was mutated in 7% (15 of 210) cases with the majority of mutations (80%, 12 of 15) located in the c-terminus of cyclin D1 (Fig 1D and Table 4). Of the 15 total CCND1-mutated cases identified, molecular data was available in 10 metastatic sites, in which the c-terminus of CCND1 was mutated in 80% (8 of 10) metastatic EMCA sites. While H&E images were not available for re-review, the majority of CCND1-mutated MSK-IMPACT advanced or metastatic EMCA cases were reportedly of endometrioid histological subtype (86.7%, 13 of 15). Additionally, one reportedly undifferentiated endometrial carcinoma also contained a c-terminal CCND1 mutation, while an endometrial serous carcinoma contained an n-terminal cyclin D1 mutation of unknown functional significance (Table 4). The combined overall findings of both TCGA and MSK-IMPACT EMCA cohorts support the notion that CCND1 c-terminal mutations are enriched in endometrial endometrioid-type carcinomas in both primary and advanced metastatic settings.
Functional effect of c-terminal CCND1 mutations in endometrial cancer
The majority of CCND1-mutated endometrial cancer cases (70%, 21 of 30 endometrial cancers, Tables 3 and 4) contained mutations in the c-terminus of cyclin D1 that may be implicated in cyclin D1 activation and gain-of-function (Tables 3 and 4). Recurrent activating mutations in endometrial cancer included: E275*, T286I (most common), P287S (second most common), D289del (Fig 1C and 1D and Tables 3 and 4). Less frequent mutations in endometrial cancer that are also likely oncogenic included: E279*, L283_294del, V290Afs*, R291dup, R291_V293del and V293del (Tables 3 and 4). Based on prior in vitro and in vivo studies, these mutations are predicted to drive cellular transformation and activation of CDK4/6 by disrupting cyclin D1 Thr-286 phosphorylation, nuclear export and proteasome-mediated degradation [12][10][11]. These findings suggest that a subset endometrial adenocarcinomas possesses a unique spectrum of somatic CCND1 mutations in which activating mutations occurring in the carboxy terminal domain of cyclin D1 are enriched, compared to other cancer types, thereby promoting cyclin D1 nuclear expression, gain of function and oncogenic activation.
Co-occurring mutations in endometrial cancer
TCGA cases exhibiting C-terminal CCND1 mutations had frequent co-occurring mutations of genes implicated in the pathogenesis of endometrial endometrioid adenocarcinomas, as the majority of CCND1-mutated cases exhibited mutations of ARID1A and members of the PI3K signaling pathway, PTEN and PIK3CA. Less frequent co-occurring mutations were seen in other genes involved in endometrial carcinogenesis such as KRAS, CTNNB1, ARID5B and PMS2 in a minority of CCND1-mutated cases. In contrast, there were no co-occurring mutations of TP53, which is frequently mutated in endometrial serous and serous-like carcinomas, or of POLE, which is altered in ultramutated endometrial carcinomas, or of MSH2, MSH6 and MLH1, which are mutated in microsatellite unstable endometrial carcinomas (S2 Fig).
C-terminal CCND1 mutations in other cancer types
Although occurring less frequently than in endometrial cancer, c-terminal activating CCND1 mutations were also identified in other cancer types, such as gastrointestinal adenocarcinoma (esophageal, stomach, colorectal, ampula of vater), squamous cell carcinoma (head & neck, lung, skin), urothelial carcinoma, cutaneous melanoma, angiosarcoma, ovarian carcinoma (low grade serous and endometrioid types) and uterine epithelioid leiomyosarcoma (Table 5). The most common activating mutations occurred in amino acids P287 and T286 as missense mutations, or as nonsense mutations in amino acids Q261, E275, E279, E280, E285, all of which are predicted to directly disrupt the phospho-degron site and/or nuclear export signal of cyclin D1, thereby resulting in nuclear accumulation and protein overexpression. The rate of overall and c-terminal activating CCND1 mutations in these non-endometrial carcinoma types was rare, at most approximately 1% in melanoma and colorectal carcinoma. The overall rate of CCND1 mutations was 1.1% in colorectal carcinoma and cutaneous melanoma (23 of 2,067 cases and 8 of 708 cases respectively) with a smaller percentage of mutations occurring specifically in the c-terminal region of cyclin D1 (0.7%, 5 of 708 cases in cutaneous melanoma and 0.4%, 9 of 2067 in colorectal carcinoma).
Discussion
Cyclin D1, encoded by the CCND1 gene, is a core cell cycle protein that promotes cellular proliferation by activating CDK4/6 kinases and progression of the G1-S phase of cell cycle. CCND1 acts as oncogene and is frequently overexpressed in a variety of cancers often via gene amplification or gene rearrangement [23][24][25]. Recently, CCND1 mutations inhibiting Thr286 phosphorylation, nuclear export and ubiquitin-mediated degradation by the proteasome have been shown to result in constitutive activation of CDK4/6 complexes, thereby promoting cellular proliferation and malignant transformation both in vitro and in vivo [6][7][8]. Corroborating its oncogenic potential, CCND1 mutations affecting Thr286 phosphorylation and inhibiting nuclear export have been reported in a few cancers [10][12][11]; however, the frequency and spectrum of CCND1 mutations have not been systematically investigated or reported across a large series of cases and across a variety of cancer subtypes. We exploited and mined the publicly available data of the Cancer Genome Atlas Project and other published exome sequencing studies to interrogate the frequency, spectrum and oncogenic effects of CCND1 mutations across of a large number and diverse set of human cancers.
Here, we demonstrate that CCND1 mutations are of very low frequency across the vast majority of cancer types (0.5%, 151 of 29,432 cases), but are predominantly enriched in endometrial endometrioid carcinomas in both the primary and metastatic settings (6.5%, 30 of 458 cases). In endometrial cancer, the majority of CCND1 mutations were located in the carboxy-terminal region of cyclin D1 (70%, 21 of 30 endometrial cancers), likely resulting in cyclin D1 oncogenic activation of cyclin D1-CDK4/6 complexes and gain of function. To our knowledge, this is the first study to systematically evaluate and report activating c-terminal CCND1 mutations across a large number of diverse cancer types. Our results are in line with two other previously published studies in which oncogenic, c-terminal cyclin D1 mutations were identified in endometrial cancer [10][11]. Ikeda et al identified CCND1 mutations in 2.3% (2 of 88) of endometrial cancer cases. Their two reported mutated cases contained a T286I mutation, which resulted in constitutive cyclin D1 nuclear accumulation and oncogenic activation [11]. Likewise, in an independent endometrial cancer cohort, Moreno-Bueno et al identified CCND1 mutations in 2.5% (3 of 119) endometrial cancer cases. Two cases demonstrated mutation at amino acid P287 (P287S and P287T), while a third case contained a 12-bp in-frame deletion that eliminated amino acids 289–292, all of which resulted in cyclin D1 nuclear expression in more than 50% of endometrial cancer cells [10]. Importantly, Benzeno et al independently confirmed that CCND1 mutations at T286, P287, D289 or deletion of amino acids 289–292 resulted in cyclin D1 nuclear retention, activation of CDK4/6 kinase and cellular transformation [12]. The rate of overall CCND1 mutations in endometrial cancer in our study is higher than that reported by Ikeda et al and Moreno-Bueno et al, likely reflecting the fact that the two prior studies sequenced a target region of cyclin D1 as opposed to next generation sequencing of the whole CCND1 gene. Our study expands the findings of Moreno-Bueno et al and Ikeda et al [10][11], in which CCND1 mutations at amino acids T286, P287, D289, deletion of amino acids 289–292 and potentially other c-terminal CCND1 mutations, such as nonsense mutation at E275*, E279*, may be important in the pathogenesis of a subset of endometrial cancers.
Importantly, the majority of TCGA cases exhibiting c-terminal CCND1 mutations had frequent co-occurring mutations of ARID1A and members of the PI3K signaling pathway, PTEN and PIK3CA, suggesting that alterations these molecular pathways are also required in addition to oncogenic nuclear cyclin D1 activation in a subset of endometrial cancers. In addition, a search of the BROAD Institute gene variant database–derived from whole exomes of healthy individuals from many ethnic or geographic settings did not reveal any of the oncogenic c-terminal CCND1 mutations or polymorphisms indentified in our study in a control population that did not harbor cancer.
Our results also corroborate the findings of Chang et al, who identified Proline 287 as a recurrent hotspot mutated amino acid of cyclin D1 in a population-scale cohort of tumor samples of various cancer types [26]. Consistent with the findings of Chang et al, Proline 287 was the most commonly mutated amino acid of cyclin D1 across all cancer types in our study. We identified mutations in cyclin D1 at Proline 287 in a diverse tumor set including melanoma, endometrial carcinoma, colonic adenocarcinoma, ovarian endometrioid adenocarcinoma, uterine leyomyosarcoma and squamous cell carcinoma of lung, oro-pharynx and skin, thus supporting the notion that CCND1 driver mutations at Pro287 may be a recurrent hotspot in cancer.
In addition, we also demonstrate high frequency of oncogenic CCND1 mutations at or surrounding amino acid Threonine 286 in the c-terminus and at or surrounding Tyrosine 44 at the n-terminus, albeit in different tumor types (i.e. carcinoma, particularly endometrial, and mantle cell lymphoma respectively). Of note, in mantle cell lymphoma, it has also been previously demonstrated that, compared to wild type CCND1, n-terminal E36K, Y44D or C47S CCND1 mutations are likely oncogenic and increased cyclin D1 protein levels through attenuation of threonine-286 phosphorylation, thereby promoting resistance to ibrutinib, which is an FDA-approved Bruton tyrosine kinase inhibitor for mantle cell lymphoma treatment [27]. Future studies are needed in order to determine whether patients, whose tumors contain oncogenic activating mutations of CCND1, would benefit from precision medicine by targeting cyclin D1-CDK4/6 kinase complexes with specific inhibitors.
In the past decade, the genomic landscape of many different types of tumors have been studied and deposited in publicly available data portals [28]. In addition, the clinical sequencing of advanced, metastatic tumors is becoming a mainstay in cancer care to guide potential treatment in mutation- and target-specific precision medicine [19]. In both clinical and preclinical research settings, a host of tools and resources can be used to interrogate specific genetic events relative to clinico-pathological characteristics of patient tumors as well as to generate hypotheses for future testing. Currently, the treatment options for advanced endometrial cancer patients are limited, and there is scant data with regards to the efficacy of palbociclib, a CDK4/6 specific inhibitor, towards endometrial cancer. In preclinical models, palbociclib or knockout of cyclin D1 had therapeutic potential against endometrial cancer cell lines and xenografts expressing the retinoblastoma protein, Rb [29][30]. Lastly, our findings implicate oncogenic c-terminal mutations of CCND1 in the pathogenesis of a subset of human cancers and provide a key resource to guide future preclinical and clinical investigations, especially with regards to the use of palbociclib or similar CDK4/6 specific inhibitors in a subset of endometrial cancers.
Supporting information
S1 Table. List of exome studies with respective cancer types and sites of origin that were mined for CCND1 mutations.
https://doi.org/10.1371/journal.pone.0199688.s001
(PDF)
S2 Table. CCND1 mutations in the mantle cell lymphoma IBIDIPS cohort.
IF del = in frame deletion; FS Ins = frame shift insertion; * = truncating mutations.
https://doi.org/10.1371/journal.pone.0199688.s002
(PDF)
S1 Fig.
Diagrams of CCND1 mutations in multiple myeloma (A), in colorectal adenocarcinoma (B) and in cutaneous melanoma (C).
https://doi.org/10.1371/journal.pone.0199688.s003
(PDF)
S2 Fig. Diagram of co-occurring mutations of genes frequently implicated in endometrial cancer in the TCGA cohort of endometrial carcinoma cases exhibiting CCND1 mutations.
Square denotes presence of gene mutation. Red bar denotes presence of c-terminal CCND1 mutation.
https://doi.org/10.1371/journal.pone.0199688.s004
(PDF)
Acknowledgments
The current study has been previously presented as an abstract and poster for the 2018 United States and Canadian Academy of Pathology annual meeting.
References
- 1. Qie S, Diehl JA. Cyclin D1, cancer progression, and opportunities in cancer treatment. J Mol Med. 2016;94: 1313–1326. pmid:27695879
- 2. Musgrove EA, Caldon CE, Barraclough J, Stone A, Sutherland RL. Cyclin D as a therapeutic target in cancer. Nat Rev Cancer. 2011;11: 558–572. pmid:21734724
- 3. Ormandy CJ, Musgrove EA, Hui R, Daly RJ, Sutherland RL. Cyclin D1, EMS1 and 11q13 amplification in breast cancer. Breast Cancer Res Treat. 2003;78: 323–35. Available: http://www.ncbi.nlm.nih.gov/pubmed/12755491 pmid:12755491
- 4. Santarius T, Shipley J, Brewer D, Stratton MR, Cooper CS. A census of amplified and overexpressed human cancer genes. Nat Rev Cancer. 2010;10: 59–64. pmid:20029424
- 5. Vose JM. Mantle cell lymphoma: 2013 Update on diagnosis, risk-stratification, and clinical management. Am J Hematol. 2013;88: 1082–1088. pmid:24273091
- 6. Alt JR, Cleveland JL, Hannink M, Diehl JA. Phosphorylation-dependent regulation of cyclin D1 nuclear export and cyclin D1-dependent cellular transformation. Genes Dev. 2000;14: 3102–14. Available: http://www.ncbi.nlm.nih.gov/pubmed/11124803 pmid:11124803
- 7. Gladden AB, Woolery R, Aggarwal P, Wasik MA, Diehl JA. Expression of constitutively nuclear cyclin D1 in murine lymphocytes induces B-cell lymphoma. Oncogene. 2006;25: 998–1007. pmid:16247460
- 8. Lin DI, Lessie MD, Gladden AB, Bassing CH, Wagner KU, Diehl JA. Disruption of cyclin D1 nuclear export and proteolysis accelerates mammary carcinogenesis. Oncogene. 2008;27. pmid:17724472
- 9. Aggarwal P, Lessie MD, Lin DI, Pontano L, Gladden AB, Nuskey B, et al. Nuclear accumulation of cyclin D1 during S phase inhibits Cul4-dependent Cdt1 proteolysis and triggers p53-dependent DNA rereplication. Genes Dev. 2007;21. pmid:18006686
- 10. Moreno-Bueno G, Rodríguez-Perales S, Sánchez-Estévez C, Hardisson D, Sarrió D, Prat J, et al. Cyclin D1 gene (CCND1) mutations in endometrial cancer. Oncogene. 2003;22: 6115–6118. pmid:12955092
- 11. Ikeda Y, Oda K, Hiraike-Wada O, Koso T, MiyasakaI A, Kashiyama T, et al. Cyclin D1 harboring the T286I mutation promotes oncogenic activation in endometrial cancer. Oncol Rep. 2013;30: 584–588. pmid:23733133
- 12. Benzeno S, Lu F, Guo M, Barbash O, Zhang F, Herman JG, et al. Identification of mutations that disrupt phosphorylation-dependent nuclear export of cyclin D1. Oncogene. 2006;25: 6291–6303. pmid:16732330
- 13. Diehl JA, Zindy F, Sherr CJ. Inhibition of cyclin D1 phosphorylation on threonine-286 prevents its rapid degradation via the ubiquitin-proteasome pathway. Genes Dev. 1997;11: 957–72. Available: http://www.ncbi.nlm.nih.gov/pubmed/9136925 pmid:9136925
- 14. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization. Genes Dev. 1998;12: 3499–511. Available: http://www.ncbi.nlm.nih.gov/pubmed/9832503 pmid:9832503
- 15. Lin DI, Barbash O, Kumar KGS, Weber JD, Harper JW, Klein-Szanto AP, et al. Phosphorylation-Dependent Ubiquitination of Cyclin D1 by the SCFFBX4-αB Crystallin Complex. Mol Cell. 2006;24. pmid:17081987
- 16. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer Statistics, 2008. CA Cancer J Clin. 2008;58: 71–96. pmid:18287387
- 17. Verma S, Bartlett CH, Schnell P, DeMichele AM, Loi S, Ro J, et al. Palbociclib in Combination With Fulvestrant in Women With Hormone Receptor-Positive/HER2-Negative Advanced Metastatic Breast Cancer: Detailed Safety Analysis From a Multicenter, Randomized, Placebo-Controlled, Phase III Study (PALOMA-3). Oncologist. 2016;21: 1165–1175. pmid:27368881
- 18. Cancer Genome Atlas Research Network, Kandoth C, Schultz N, Cherniack AD, Akbani R, Liu Y, et al. Integrated genomic characterization of endometrial carcinoma. Nature. 2013;497: 67–73. pmid:23636398
- 19. Zehir A, Benayed R, Shah RH, Syed A, Middha S, Kim HR, et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat Med. Nature Publishing Group; 2017;23: 703–713. pmid:28481359
- 20. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2: 401–4. pmid:22588877
- 21. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6: pl1. pmid:23550210
- 22. Chakravarty D, Gao J, Phillips S, Kundra R, Zhang H, Wang J, et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precis Oncol. American Society of Clinical Oncology; 2017; 1–16. pmid:28890946
- 23. Diehl JA. Cycling to cancer with cyclin D1. Cancer Biol Ther. 1: 226–31. Available: http://www.ncbi.nlm.nih.gov/pubmed/12432268 pmid:12432268
- 24. Hardisson D. Molecular pathogenesis of head and neck squamous cell carcinoma. Eur Arch Otorhinolaryngol. 2003;260: 502–8. pmid:12736744
- 25. Lukas J, Jadayel D, Bartkova J, Nacheva E, Dyer MJ, Strauss M, et al. BCL-1/cyclin D1 oncoprotein oscillates and subverts the G1 phase control in B-cell neoplasms carrying the t(11;14) translocation. Oncogene. 1994;9: 2159–67. Available: http://www.ncbi.nlm.nih.gov/pubmed/8036001 pmid:8036001
- 26. Chang MT, Asthana S, Gao SP, Lee BH, Chapman JS, Kandoth C, et al. Identifying recurrent mutations in cancer reveals widespread lineage diversity and mutational specificity. Nat Biotechnol. 2016;34: 155–163. pmid:26619011
- 27. Mohanty A, Sandoval N, Das M, Pillai R, Chen L, Chen RW, et al. CCND1 mutations increase protein stability and promote ibrutinib resistance in mantle cell lymphoma. Oncotarget. Impact Journals, LLC; 2016;7: 73558–73572. pmid:27713153
- 28. Chandran UR, Medvedeva OP, Barmada MM, Blood PD, Chakka A, Luthra S, et al. TCGA Expedition: A Data Acquisition and Management System for TCGA Data. Ebrahimie E, editor. PLoS One. 2016;11: e0165395. pmid:27788220
- 29. Tanaka T, Terai Y, Ashihara K, Fujiwara S, Tanaka Y, Sasaki H, et al. The efficacy of the cyclin-dependent kinase 4/6 inhibitor in endometrial cancer. Castresana JS, editor. PLoS One. Public Library of Science; 2017;12: e0177019. pmid:28472136
- 30. Dosil MA, Mirantes C, Eritja N, Felip I, Navaridas R, Gatius S, et al. Palbociclib has antitumour effects on Pten- deficient endometrial neoplasias. J Pathol. 2017;242: 152–164. pmid:28349562