UBC Faculty Research and Publications

New classification of endometrial cancers: the development and potential applications of genomic-based… Talhouk, A.; McAlpine, J. N Dec 13, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-40661_2016_Article_35.pdf [ 1.13MB ]
JSON: 52383-1.0368589.json
JSON-LD: 52383-1.0368589-ld.json
RDF/XML (Pretty): 52383-1.0368589-rdf.xml
RDF/JSON: 52383-1.0368589-rdf.json
Turtle: 52383-1.0368589-turtle.txt
N-Triples: 52383-1.0368589-rdf-ntriples.txt
Original Record: 52383-1.0368589-source.json
Full Text

Full Text

REVIEW Open AccessNew classification of endometrial cancers:the development and potentialapplications of genomic-based classificationin research and clinical careA. Talhouk1 and J. N. McAlpine2*AbstractEndometrial carcinoma (EC) is the fourth most common cancer in women in the developed world. Classification ofECs by histomorphologic criteria has limited reproducibility and better tools are needed to distinguish these tumorsand enable a subtype-specific approach to research and clinical care. Based on the Cancer Genome Atlas, tworesearch teams have developed pragmatic molecular classifiers that identify four prognostically distinct molecularsubgroups. These methods can be applied to diagnostic specimens (e.g., endometrial biopsy) with the potential tocompletely change the current risk stratification systems and enable earlier informed decision making. Theevolution of genomic classification in ECs is shared herein, as well as potential applications and discussion of theessential research still needed in order to optimally integrate molecular classification in to current standard of care.Keywords: Endometrial carcinoma, Histotype, The Cancer Genome Atlas (TCGA), Risk stratification, Prognosis, POLEmutations, Mismatch repair deficiencies, p53, Molecular classificationBackgroundCancer care in the last decade has featured a concertedmove towards the personalization of patient care, oftencalled precision medicine. In the field of cancer, this hasmeant a progression from broad categorization of tumorsby anatomic site, to distinguishing subgroups by histomor-phology, and more recently defining tumors by molecularfeatures. This evolution has not happened over night andpace of change has varied by tumor site. Paradoxically,despite endometrial cancer being the most common gyne-cologic malignancy in women in Canada and the UnitedStates [1, 2] and the 6th most common cancer in womenglobally [3], research and clinical advancement havearguably lagged as compared to other cancers. Thismay be because over 75% of women diagnosed withendometrial cancer have early stage disease (stage I or II)and favorable outcomes (5-year overall survival 75–90%)[4–6]. However, for those women who recur or for thosewho present with more advanced disease, response ratesto conventional chemotherapy are low and clinical out-comes are extremely poor [7–10].Renewed research focus on this disease site hasbeen prompted by a dramatic increase in incidenceobserved in the developed world [2, 11–13]. Inaddition, there has been frustration with contemporarypractice, in part due to inconsistent EC histomorphologiccategorization, imprecise risk stratification, and diversetreatment strategies. Multidisciplinary panel recommen-dations on management of ECs [14] have emerged in aneffort to make treatment (surgery, chemotherapy, radio-therapy, surveillance) more consistent. Multiple reviewson state of the art care of EC’s have been published, andincreasingly the repercussions of treatment on patientquality of life are being assessed in addition to survivalparameters [6, 15–20]. Attention to this balance oftreatment and sequelae may be even more essential inthis disease site as there is concern that many womenare likely over-treated or under-treated.* Correspondence: jessica.mcalpine@vch.ca2Department of Gynecology and Obstetrics, Division of GynecologicOncology, University of British Columbia, 2775 Laurel St. 6th Floor,Vancouver, BC, CanadaV5Z 1M9Full list of author information is available at the end of the article© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 DOI 10.1186/s40661-016-0035-4There has been a call for the incorporation of molecu-lar features in to both classification and risk determin-ation of ECs in order to better assess the biologicalbehavior of an individual’s disease and ultimately to im-prove treatment decisions and outcomes [21, 22]. Theobjective of this review is to focus on the new genomicframework used to categorize endometrial carcinomas.Herein we describe the evolution of molecular classifica-tion systems and how genomic characterization will im-pact both research approach and clinical managementfor this disease.Historical/Pathogenetic Classification ofEndometrial CancerThirty years ago, Bokhman hypothesized there weretwo pathogenetic types of endometrial carcinomasdriven by very different metabolic and endocrinesignals [23]. Type 1 is more common (~70–80%), consist-ing of endometrioid, low grade, diploid, hormone-receptorpositive tumors that are moderately- or well-differentiatedand more common in obese women. Patients presentingwith Type 1 tumors tend to have localized disease con-fined to the uterus and a favourable prognosis. In contrast,Type 2 tumors (20–30%) are more common in non-obesewomen, of non-endometrioid histology, high-grade, aneu-ploid, poorly differentiated, hormone receptor negativeand associated with higher risk of metastasis and poorprognosis. While this historical system of taxonomy hasbeen useful, substantial heterogeneity within and overlapbetween Type I and II cancers is now recognized. Type Iand Type II designation has never been part of the formalstaging nor risk stratification, and thus has no clinicalutility beyond providing a conceptual framework forunderstanding endometrial cancer pathogenesis.Endometrial classification by histomorphologyand current systems of risk stratificationTumor grade and histologic subtype assessment aresubjectively assigned according to appearance under themicroscope and predefined pathologic criteria. Nuclear fea-tures and the proportion of solid tumor vs. identifiableglands defines grade 1–3. Histologic subtype is assigned bymorphologic criteria and often aided by immunostains.Pathologic accuracy is hampered by poor diagnostic repro-ducibility, especially in the case of high-grade subtypes (e.g.grade 3 endometrioid, serous). Studies describe inter-ob-server disagreement or lack of consensus on histologicsubtype diagnosis in one-third or higher of ECs [24–27]. The overall kappa statistics for FIGO grade assign-ment between pathologists is 0.41–0.68, indicative ofonly moderate levels of inter-observer agreement [24,28]. Agreement between diagnostic specimens andfinal hysterectomy is also limited [29–32]. In short,histologic classification is not accurate or preciseenough to effectively triage patients into optimal treat-ment groups.Endometrial carcinoma has been a surgically stageddisease since 1988. Surgery traditionally involves hyster-ectomy with bilateral salpingo-oophorectomy +/− lymphnode dissection or sampling and omentectomy with severalsafe options in surgical approach [14, 33–35]. Extentof staging may vary according to patient age, comorbidi-ties, cancer histology, grade, disease distribution, surgeonpreference and institutional practice. Surgery alone is typ-ically sufficient to cure early-stage EC [14, 36, 37], how-ever, it is recognized that tumors with ‘high-risk’ featureshave a high likelihood of recurrence and adjuvant treat-ment (radiation and/or chemotherapy) is recommended[8, 16, 38, 39]. The major challenge is in distinguishing thefeatures that comprise ‘low-‘, ‘intermediate-‘, and ‘high-risk’ disease in ECs. Multiple different risk predictiveclinical models have been developed to guide treatment[14, 37, 40–48]. These have evolved with new FIGOstaging and through interpretation of large clinicaltrials, however all incorporate the key pathologicalparameters of histotype, grade, and stage. As mentionedpreviously, the reproducibility of both histotype andgrade have been demonstrated to be poor in EC’s[24, 26, 27], thus two of three major criteria for riskgroup assignment which directly impacts recommenda-tions for adjuvant treatment have limited reproducibil-ity. Understandably, this makes it challenging toconfidently make treatment decisions. We know thatsome women are undertreated who could have bene-fited from aggressive surgery, chemotherapy and/or ra-diation, and many may be overtreated having beencured by surgery alone.The adequacy of risk stratification systems in EChave recently been compared and challenged [22, 49].There are five major risk stratification systems in EC,of which the modified European Society of MedicalOncologists (ESMO) classification was demonstratedto best discriminate for recurrence and nodal metasta-ses in apparent early stage disease [49]. However, noneof the existing schemes were deemed highly accurate.In addition, all current systems stratify women basedon pathologic data obtained after surgical staging(stage is a component of risk assignment). There isgreat need to obtain earlier and more biologically in-formative data from EC tumors that could assist inplanning the optimal course of treatment for the indi-vidual. In addition, diagnostic tools that could object-ively and consistently categorize ECs into distinctsubgroups would enable stratification of clinical trialsand study of treatment efficacy within biologically ‘like’subgroups. Stemming from clinical need and a recog-nized inadequate/unsustainable system a call was madefor the integration of molecular features.Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 2 of 12A new genomic era: molecular classification ofendometrial carcinomasSeveral research teams have defined immunohistochemicaland/or mutation profiles to aid in distinguishing EC sub-types [50–58]. In one series, a set of seven immunohisto-chemical markers was able to improve the distinctionbetween high-grade EC histotypes [28] and more recently,another team demonstrated a nine protein panel improvedidentification of both low and high-grade EC subtypes [57].Sequencing has enabled further improvement, with a nine-gene panel, demonstrating distinct mutational profiles forthe major EC histotypes [52]. Molecular data has also beenused to further stratify risk categories; using gene expres-sion profiling and copy number analysis to determine riskof recurrence [59, 60], even in apparent low stage disease[61]. Molecular characterization has also been pursued forpotential therapeutic targets in EC, focusing on frequentlymutated pathways such as PI3K/PTEN/AKT/mTOR. Fur-ther work is needed to define molecular biomarkers thatmore accurately reflect tumor susceptibility [62–66].The most comprehensive molecular study of ECs todate has been The Cancer Genome Atlas (TCGA) project,which included a combination of whole genome sequen-cing, exome sequencing, microsatellite instability (MSI)assays, and copy number analysis [67]. Molecular informa-tion was used to classify 232 endometrioid and serousendometrial cancers into four groups - POLE ultramu-tated, MSI hypermutated, copy-number (CN) low, andCN high - that correlate with progression-free survival.The ultramutated POLE subgroup was a novel findingfrom the TCGA, and generated interest due to its veryfavorable outcomes even within high-grade tumors. InTCGA, ultramutated cases were characterized by POLEexonuclease domain mutations (EDM), a high percent ofC > A transversions, a low percent of C > G transver-sions, as well as more than 500 SNVs. POLE encodes themajor catalytic and proofreading subunits of the Polε(Polymerase Epsilon) DNA polymerase enzyme complexresponsible for leading strand DNA replication. The exo-nuclease proofreading function and the high fidelity in-corporation of bases by POLE ensures a low mutationrate in the daughter strand. In ECs, POLE EDMs aremostly found in hotspot regions with V411L and P286Rbeing the most common mutations. Substitutions in DNApolymerases were shown to inactivate or suppress proof-reading abilities, thus causing increased replicative errorrates and resulting in the ultra-mutated phenotype. In theTCGA, whole genome or exome sequencing was usedto assess POLE status. Other series have subsequentlyassessed POLE status using more focused methods includ-ing Sanger sequencing [68, 69], gene panels [69–71], digitalPCR [72–74] or functional assays [75] and confirmedvery favourable outcomes for women with POLE aber-rant ECs.TCGA also described a molecular subgroup that ex-hibited microsatellite instability (MSI). MSI arises fromdefects in post-replicative DNA mismatch repair system.In the TCGA, MSI was determined by a panel of fourmononucleotide repeat loci (polyadenine tracts BAT25,BAT26, BAT40, and transforming growth factor receptortype II) and three dinucleotide repeat loci (CA repeats inD2S123, D5S346, & D17S250) in addition to the recom-mended markers from the National Cancer Institute [76],tumor DNA was classified as microsatellite- stable (MSS) ifzero markers were altered, low level MSI (MSI-L) if one totwo markers (less than 40%) were altered and high levelMSI (MSI-H) if three or more markers (greater than 40%)were altered. Mismatch repair deficiencies can resultfrom i) an inherited cancer syndrome (e.g., Lynch), ii)acquired/somatic mutations or iii) epigenetic events e.g.methylation of one of the genes involved in mismatchDNA repair, most commonly MLH1.Finally TCGA distinguished a distinct molecular sub-group by copy number analysis. Copy number was deter-mined using Affymetrix SNP 6.0 microarrays using DNAoriginating from frozen tissue. Hierarchical clusteringidentified significantly reoccurring amplifications ordeletions regions and a ‘copy number (CN) high’ sub-group. All remaining samples that did not belong tothe POLE ultramutated group, the MSI group, or theCN high group, were termed CN low. The appeal ofobjective molecular categorization of new EC cases into one of four prognostic subgroups was immediatelyapparent. However, methodologies used for the TCGAstudy were costly, complex and unsuitable for widerclinical application.Two research teams, including our own, have subse-quently developed more pragmatic methodologies toevaluate molecular features of ECs, working in standardformalin-fixed paraffin-embedded tissue. These methodsdo not identify molecular subgroups that are identical toTCGA but do recapitulate the four survival curves ob-served in TCGA [69, 71, 73, 77] (Fig. 1). Stelloo et al.[69, 71] used a combination of TP53 mutational testingand p53 IHC to determine p53 status obtained from se-quencing as a surrogate for CN high TCGA subgroup.The promega MSI analysis system was used to determineMSI status. For tumors exhibiting low levels of instabilityor from which extracted DNA quality was poor, immu-nohistochemistry for mismatch repair (MMR) proteins(MLH1, MSH2, MSH6, and PMS2) was performed. POLEEDM hotspot mutations were identified by Sanger sequen-cing. This team also tested for hotspot mutations (159)across 13 genes (BRAF, CDKNA2, CTNNB1, FBXW7,FGFR2, FGFR3, FOXL2, HRAS, KRAS, NRAS, PIK3CA,PPP2R1A, and PTEN). Testing ultimately yielded fourmolecular subgroups: group 1 - p53 (mutation identified),group 2- MSI, group 3 –POLE (POLE EDM identified), andTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 3 of 12finally group 4 –NSMP, a group with ‘no specific molecularprofile’ (Fig. 1a). Tumors with insufficient tissue to performall molecular testing were not classified and tumors withmore than one molecular feature, constituting 2–3% of thecohort, were also not classified. Due to this exclusion,the order of mutational testing was irrelevant. This researchteam initially assessed ECs from the PORTEC3 trial (n =116), with known high risk features. Recurrence-free sur-vival and time to distance metastasis were assessed withinthe four molecular subgroups. They observed that patientsbelonging to the POLE and the MSI subgroups showedsimilar and much better survival outcomes in comparisonto the p53 mutant group and the NSMP group which ex-hibited worse recurrence and distance metastasis outcomeseven within the endometrioid histology cases. Differencesin survival patterns relative to the TCGA results were at-tributed to a greater proportion of high-risk features in thePORTEC 3 cohort.The Leiden/TransPORTEC group has since applied thesame series of molecular tests to a larger, more diverse co-hort [71]. However, survival analysis and assessment ofprognostic ability was restricted to endometrioid subtypeand stage 1 tumors of patients with intermediate clinicalrisk. Within this very specific group, the observed out-comes associated with each molecular subgroup moreclosely mirrored TCGA.Our research team has also developed a molecularclassification system that uses practical methodologies toassign ECs to one of four molecular subgroups with dis-tinct survival outcomes. We have followed the Institute ofMedicine (IOM) guidelines for the development of ‘omicsbased tests [78], initially exploring 16 models in a ‘discov-ery’ cohort (n = 141) [73], next locking down sequence oftesting and methods to a single model termed ProMisE(Proactive Molecular Risk Classifier for EndometrialCancer) on a new ‘confirmation’ cohort (n = 319) [77,79] to prove feasibility and confirm the associationwith outcomes/prognosis, and finally testing in a large ‘val-idation’ cohort (n =~500) of ECs from collaborators at theUniversity of Tübingen (Germany). Molecular decision treeanalysis for ProMisE is outlined in Fig. 1b. Specific method-ologies include immunohistochemistry (IHC) for the detec-tion of the presence/absence of two mismatch repair(MMR) proteins: MSH6 and PMS2. This identifies ‘MMR-D’ (deficient) subgroup. Cases are then sequenced usingdigital PCR to identify POLE exonuclease domain muta-tions (‘POLE EDM’). Finally, cases are assessed using IHCfor p53 (wild type vs. null or missense mutations; ‘p53wt’and ‘p53abn’, respectively). We have demonstrated thatwomen within each molecular subgroup have clinicopath-ological characteristic that have consistently been shownto be typical of that group. For example, the p53 abn sub-group usually encompasses the highest proportion of highgrade, advanced stage, non-endometrioid histotypes andarises in older, thinner women. Similarly, the emergingphenotype of women whose EC harbor POLE EDMs is ofparticular interest since it generally includes younger,thinner and with surprisingly aggressive pathologic fea-tures (large proportion of grade 3 tumors, many with deepmyometrial invasion and LVSI) yet consistently exhibit fa-vorable outcomes. The MMR-D subgroup have very simi-lar ‘uterine factors’ (clinicopathologic features in theuterus itself) [48] to the POLE subgroup, i.e. a comparableproportion of high grade tumors and deep myometrial in-vasion and LVSI, yet they have worst observed outcomesof any group next to p53abn [77, 79]. On multivariableanalysis, ProMisE molecular subgroup assignment main-tained its association with overall survival (OS), progres-sion free survival (PFS) and recurrence free survival (RFS)even after correction of other clinicopathologic parametersof known prognostic significance available at time of diag-nosis/collection of diagnostic specimen for molecular ana-lysis (e.g., age, BMI, grade, histotype but not stage).Both ProMisE (across all tumors tested), and the Leidenclassifier (within the intermediate-risk group examined)demonstrate comparable risk discriminatory ability to theESMO risk stratification system. Furthermore when clinicaland pathological features were integrated with molecularfeatures they resulted in improved risk stratification.Through evaluation of the collective cohort (discovery +Fig. 1 Schematic of the a Leiden/TransPORTEC and b ProMisE/Vancouver molecular classification systems including testing performed,molecular subgroups identified, and by what criteria cases would beconsidered unclassifiableTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 4 of 12confirmation + validation cohorts = ~ 1000 ECs) we plan toevaluate which key clinicopathological parameters can addvalue to molecular classification giving high priority tothose features available at time of diagnosis (e.g., age, BMI).Our goal has consistently been to develop a molecularclassification tool that could be applied to diagnosticspecimens (endometrial biopsy or curettage) and thereforeinform treatment at the earliest time point. Biologicallyrelevant information about an individual’s tumor couldguide surgical urgency and aggressiveness, fertility orhormonal function sparing management options, adjuvanttherapy, and/or surveillance schedules. We have demon-strated high concordance between ProMisE molecular clas-sification in diagnostic vs. final hysterectomy samples, farsuperseding concordance of grade, or histotype as assignedon original pathology reports or within or between reviewsby expert gynecologic cancer pathologists [80]. The Leidenteam has also shown high concordance of molecular tumoralterations between pre-operative curettage specimensand final hysterectomy specimens (13 gene panel andMSI assay) [81] and a multicenter, prospective trial inHolland is in process to see if surgical management canbe improved [82]. As diagnostic specimens are fixedimmediately (in contrast to a hysterectomy specimenthat may sit for hours in an operating room before pro-cessing in pathology), the quality of DNA extracted andfixation for IHC is high. We believe one of the most ex-citing aspects of molecular classification and what willbe most impactful in directing care for women with ECwill be this capability of determining earlier prognostic(and possibly predictive) information.Ultimately, integration of molecular classification byeither method into current practice, as performed ondiagnostic specimens or final hysterectomy, will needto be studied in the context of a prospective clinicaltrial; comparing survival outcomes, quality of life andhealth economic implications to conventional/historicalstandard of care.Challenges with molecular classification: keycomponentsThe Leiden/TransPORTEC and Vancouver/ProMisE prag-matic molecular classification systems incorporate the sameintegral components: identification of ECs with mismatchrepair deficiency/microsatellite instability, POLE exonucleasedomain mutations and aberrant p53. Similarities anddifferences are shown in Fig. 1. Prognostic strength ofmolecular classification is at least equivalent to otherclinicopathological features or risk stratification systemsbut offers the advantage of objective results (e.g., presenceor absence of a protein or mutation). We believe these keymolecular components are unlikely to be outperformed byany single clinicopathological parameter or biomarker. Not-ably, as yet none of the additional immunohistochemicalmarkers we have tested across our endometrial cancer caseshave outperformed ProMisE. Although we and others areinvestigating the immune landscape and specific immuno-histochemical biomarkers within the context of these majormolecular subgroups these studies will not be covered inthis manuscript. Should any parameter improve the abilityto discern outcomes and guide management beyondthe ProMisE or Leiden molecular classification, theycan be incorporated into future algorithms. Herein, wefocus on some of the major challenges and consider-ations for future implementation of molecularclassification.MMR/MSIThere are different techniques for the identification ofmismatch repair deficiency [76, 83–87]. Both TCGAand the Leiden series use microsatellite instability (MSI)assays. These have primarily been utilized in research, notclinical practice settings (there are no FDA-approved MSItests) and require DNA extraction from tumor as well asnormal tissue or blood for comparison. ProMisE tests forthe presence of two mismatch repair proteins (MSH6,PMS2) by immunohistochemistry and we have shown highconcordance between MMR IHC and MSI assay methods[83]. IHC staining for MMR and interpretation is routinefor most pathology laboratories. Unfortunately, althoughhistomorphologic surrogates for MMR deficiency or Lynchsyndrome have been explored (e.g., tumor infiltrating and/or peritumoral lymphocytes, dedifferentiated histology,lower uterine segment origin) [88, 89], as yet they have notproven to be equivalent to molecular confirmation.Although all MMR deficiencies are often groupedtogether, for inherited mutations (Lynch syndrome),the lifetime risk and age of penetration of Lynch-associatedcancers can vary substantially according which gene is aber-rant [90]. This may impact recommendations regarding thetiming of screening or intervention e.g., lower lifetime riskand later average age of penetration for individualswith aberrant MSH6 [90–93] might enable delay ofrecommended risk reducing surgery as compared toother Lynch mutations.Prognostic and predictive implications of mismatch re-pair may also vary according to specific MMR gene muta-tion or protein loss identified. It has been hypothesized thatepigenetic/methylation events in mismatch repair likelyhave different implications on tumor characteristics andclinical outcomes than germline defects e.g. an age-relatedsomatic event would not be expected to promote the devel-opment of tumor that is equivalent to one arising in ayoung individual harboring a germline mutation. Immuneenvironment, intrinsic biologic behavior, toleration of adju-vant therapy/response to cell injury may vary significantlyin these individuals. This may partially explain the rela-tively wide range of response to immunotherapy withinTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 5 of 12MMR-D cases. At present, all mismatch repair deficien-cies are lumped together but further interrogation ofthese differences (e.g. subgroups of subgroups) is war-ranted. Recently, over 1000 women with EC had their tu-mors evaluated for microsatellite instability, MLH1methylation, and MMR protein expression as part of acombined NRG Oncology/Gynecologic Oncology GroupStudy (GOG210) [94]. Categories of normal mismatch re-pair, epigenetic defect and probable mutation (somatic orgermline) were compared to clinicopathologic variablesand clinical outcomes in the trial cohort. Even with thislarge number of cases, these three broad categories ofMMR status were not shown to be associated with PFS orDSS. Univariate analysis did suggest potentially worse PFSfor women whose tumors had epigenetic defects (trend, p= 0.1) but this association was not maintained afteradjusting for other factors, including the highly relevantparameter of age in this cohort. In addition, the authorsobserved a trend to improved PFS in tumors with MMRmutations and a suggestion that these patients receivedgreater benefit from adjuvant chemotherapy compared towomen with normal mismatch repair. Similar results forprobable germline/Lynch syndrome mismatch repairdeficient tumors were observed in a smaller series of221 ECs, with no prognostic nor predictive associa-tions noted in the tumors with methylation events[95].POLESeveral research teams have characterized POLE mutatedtumors by histomorphology and immune environment[96–101]. Obvious clinical implications for tumors withsubstantial immune infiltrates include selection foranti-PD-1 therapy. However, the highly favorable out-comes observed in women with POLE mutated tumorswould suggest that costly targeted therapy might better bereserved for the very rare cases of recurrent or advanceddisease [102, 103]. POLE somatic mutations are found inless than 10% of endometrial carcinomas and recurrenceis seldom observed; thus, it has been difficult for a singlestudy to be adequately powered to determine optimalmanagement of women whose tumors harbour this mo-lecular feature. Adjuvant treatment is commonly adminis-tered due to the frequency of ‘high-risk’ features in ECswith POLE EDMs (e.g., relatively high frequency of grade3, deep myometrial invasion, LVSI) but whether this isover treatment of women who would do well based ontheir POLE genotype alone or whether treatment isneeded and favorable outcomes are secondary to exquisitesensitivity to DNA damaging agents in these tumors is asyet unclear.The paradox of observed aggressive histopathologicfeatures but excellent survival outcomes may in part beexplained by the high neoantigen load and immune richmicroenvironment in tumors with POLE EDMS (and toa lesser degree, also described in MMR-D tumors).Both sequencing and functional assays currently employedfor POLE mutation testing are more costly than IHC andutilize methods that require a skilled team to perform andinterpret. We, and others, continue to search for surro-gates for POLE sequencing. Although the clinical andpathological phenotype of women with POLE mutatedtumors is beginning to be characterized; on average youn-ger, lower BMI, high proportion of grade 3, LVSI+, pre-dominantly endometrioid, and low stage, these parametersoverlap with other molecular subgroups. At present thereis no single pathognomonic surrogate for this feature.p53The mutational spectrum of TP53 mutations within ECswas recently described in Schultheis et al. [104], both inthe context of histotype and across TCGA molecularsubgroups. This study confirmed the very high propor-tion (91%) of TP53 mutations in the ‘CN high’ TCGAcategory but also seen in 35% of the POLE genomic sub-group. No clinical correlative data was provided withtheir paper but our series and others confirm the highlyfavorable outcome of POLE mutation carriers even withthe identification of other mutations traditionally associ-ated with high risk disease. The order of ourcategorization: identification and removal of POLE sub-group prior to p53 stratification thus seems to be ofgreat importance (see tumors with >1 molecular featurebelow). Also described in this series was the presence offrameshift or nonsense TP53 mutations (22% of TP53mutant subset) of which they acknowledge would yielddifferent IHC results (loss or IHC score 0) than missensevariants (IHC score 2). Identification of both aberrantstates is essential. Our team, in collaboration with othersis in the process of further characterizing both TP53 mu-tational and IHC status in ECs in order to better guideinterpretation in this disease site.Tumors with more than one molecular featureBoth Talhouk et al. [77, 79] and Stelloo [71] et al.describe approximately 2-3% of endometrial tumorshaving more than one of the key molecular featuresdescribed. Reported frequency of post-replicationPOLE proofreading defect and a DNA mismatch re-pair defects varies in the literature, but in serieswhere co-occurrence is higher, this has been attrib-uted to somatic MMR mutations which may be sec-ondary to the ultra-mutated POLE phenotype. [67,70, 98]. Similarly, it is perhaps not surprising that inboth the POLE and MMR-D subgroups of ECs withhigh mutational loads, tumors may also harbourTP53 mutations (as evidenced by either sequencing,or complete loss or overexpression of p53 protein onTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 6 of 12IHC) [71, 77, 79, 104]. The order of testing for mo-lecular classification is therefore critically important.Determination of POLE status prior to p53 testingwill categorize a given EC as POLE EDM. Favorableoutcomes are therefore anticipated for that individual,and indeed for cases reported thus far with dual fea-tures, that has been observed [77]. We believe testingfor MMR-D first is still valid, as that information is ar-guably more actionable than POLE status (referral forhereditary testing, consideration of immunotherapy)which is not currently integrated into treatment algo-rithms. Ultimately, distinguishing between what are likelypassenger mutations or late events without functional con-sequence as compared to mutations that define biologicbehavior is essential. Molecular classification tools thatutilize large gene panels may detect a plethora of coexistingmutations in POLE EDM ECs and need to be interpretedwith caution e.g., discovery of BRCA1 or BRCA2 mutationin a POLE mutated EC may not indicate homologous re-combination deficiency / PARPi efficacy [105].Clinical outcomes may be harder to discern betweenECs demonstrating both MMR deficiency and p53 muta-tions and the ‘best’ categorization of these tumors remainsto be determined. At present, molecular classification willfirst identify the MMR deficiency at least enabling patientsto be referred for hereditary counselling and providing op-portunities in genotype specific clinical trials.Genotype-phenotype interplayGenotype-phenotype interactions have been appreciatedand characterized in recent years. Although not the focusof this review, we will take this opportunity to describeone highly relevant example.It is now appreciated that PTEN loss has differentprognostic implications in lean vs obese individuals.Mutations in the central relay pathways of insulin signals(phosphatidylinositol 3-kinase (PI3K) pathway includingmutations specifically in PIK3CA, PIK3R1 and PTEN) areextremely common in ECs yet prior studies on the prog-nostic significance of PTEN mutations had markedly dis-cordant results. Westin et al. stratified cases by body massindex (BMI) revealed improved progression free survival inobese (BMI >30) women with endometrioid endometrialcarcinoma suggesting an interaction between metabolicstate and genetics [106]. Subsequently, a constellation of‘obesity related’ genes are observed to be upregulated withincreasing BMI among endometrioid carcinomas in theTCGA cohort [107], and different targets for treatmentwere suggested in obese vs non-obese individuals [108].Given the global epidemic of obesity and associated‘metabolic syndrome’, this clinical context is essential toknow in guiding clinical management and in research/interpretation of data. In our own series, for example,we anticipate, that further stratification of cases withinthe p53 wt subgroup (and possibly within MMR-D) byBMI status may refine prognosis further. We are in theprocess of examining the interaction of PTEN and BMIwithin the ProMisE molecular subgroups across all ofour evaluable cohorts.Rare histotypes and diversity within tumorsThe role of molecular classification in rare histotypes ofendometrial carcinoma has not been determined. TheTCGA was restricted to cases of endometrioid and seroushistology, however, the TransPORTEC cohorts and ourown series included other histologies; 15% clear cell, and acombination of 6% clear cell, carcinosarcoma, undifferenti-ated, and mixed, in the cohorts respectively [69, 77, 79].Fundamental features of the immunophenotype for dedif-ferentiated, clear cell, and mixed carcinomas have been re-ported [50, 54, 109–111]. Assessment of mixed tumorsshow that despite morphologic differences/mimicry, themajority of molecular aberrations are shared across thetumor [112]. Thus the application of ProMisE or Leidenclassification systems to these cancers may be of value. In-deed in the small number of non-serous, non-endometrioidcases studied thus far, histotypes were distributed acrossthe molecular subgroups (not confined to p53 abn sub-group). We anticipate there will be deeper characterizationof unique genomic categories; e.g., dedifferentiated carcin-omas within p53 wt subgroup with mutations in the SWI/SNF pathway.Intratumoral heterogeneity in EC has been described[113, 114], and might be predicted to weaken the utilityof ProMisE. However, in the cases examined, althoughsingle nucleotide variations and copy number analysisrevealed some diversity between anatomic sites withinan individual (at time of diagnosis) the ProMisE molecularsubgroup categorization was concordant across all tumorsites (6–14 anatomic sites examined per individual) [114].We have reported on a case of discordant ProMisEcategorization between a diagnostic endometrial biopsyand final hysterectomy specimen in an individual with adedifferentiated endometrial carcinoma [80]. This wassecondary to concurrent low grade and high grade areaswithin the endometrium and myometrium where mismatchrepair profiles differed. In rare cases, in which diversetumor morphology is observed it may be that morethan one area needs to undergo molecular testing. Cer-tainly, gross and microscopic assessment of endometrialcancers by pathologists will need to continue just as rele-vant post staging data on metastases may be weighed inmanagement. Successful integration of molecular classifi-cation will require addressing all of these issues over time,but in the interim, we anticipate a mix of current practice(histomorphologic categorization) and molecular tools forassessment of newly diagnosed ECs.Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 7 of 12ConclusionsWe have harboured too long in a system of irreprodu-cible categorization of endometrial carcinomas, incon-sistent management within and across cancer centers,and inappropriate research investigations that groupeddiverse tumors for study, making advances in researchand clinical management slow or impossible in this diseasesite. It is essential that biologically relevant molecularfeatures are assessed and considered for categorization oftumors, and in deciding surgical management and adjuvanttherapy. This does not require abandonment of clinicopath-ologic parameters, many of which have been demonstratedto maintain prognostic relevance even in the post-TCGAera, but rather not to rely on them as the only or mostimportant feature to guide management.We have shown that in the hands of two independentresearch teams molecular classification of endometrialcarcinomas is feasible, and identifies four prognosticallydistinct subgroups. Historical segregation of Type I(mostly CN low, p53 wt cases) and Type II ECs (mostlyCN high, p53 abn subgroups) is inadequate and do notaccount for the approximately 30% of cases that areMMR-D or POLE EDM. All components of the molecularclassifier together can be achieved at a cost* comparable toother commonly utilized clinical assays in cancer care.At minimum, this system provides objective reproduciblecategorization of EC’s. Familiarity with MMR and p53 IHCtesting and interpretation lends to rapid adoption in anypathology department. The reproducibility of ProMisEacross Canadian cancer centers is currently beingevaluated.Additional benefits of molecular classification includeearly identification of women who may have an inheritedgenetic syndrome (Lynch) who would benefit from add-itional screening or interventions for other Lynch-associated cancers or in whom specific therapies for theirTable 1 Potential changes in practice through molecular categorization*Features that have been historically used in risk classification or are considered prognostic markers in other seriesTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 8 of 12endometrial carcinomas may be more effective. For youngwomen with EC considering delay of hysterectomy forfertility reasons (e.g., progesterone therapy), molecularclassification of her diagnostic endometrial specimencould help guide management as either MMR-D (de-pending on germline results post hereditary cancer re-ferral) or p53 abn categorization would discourage aconservative approach. It is still unclear how knowingthe POLE mutation status within an individual’s EC willimpact her clinical management, as favorable outcomesobserved in these individuals may be either independentor secondary to increased sensitivity to DNA damagingagents (chemotherapy, radiation), and withholding treat-ment cannot yet be advised. Plausibly, women with p53abn tumors with higher association of metastatic diseaseand aggressive clinical course would be recommended toundergo more comprehensive surgical staging and closersurveillance. Conversely, biologically indolent tumorsmay be cured by simplified surgery alone and perhapsspared toxic treatment and managed by communitygynecologists.We are at an exciting juncture, but aware of the manyquestions still remaining (Table 1). Through clinical trials,we need to determine how molecular classification can bebest integrated in to current clinical care and how will itimpact outcomes. What, if any, additional parameters canbetter inform management? Interrogation of genotypic andphenotypic features may provide additional prognostic andpredictive information. These can now be explored withinthe context of the four major molecular categories of tu-mors, and even within molecular subgroups. How reliableare IHC surrogates for mutational data in endometrial can-cer? Characterization of p53 and other markers, as has beenachieved in ovarian cancer [115] is needed for this diseasesite. Can molecular classification help interpret the naturalhistory and direct management of cases that have historic-ally been a great challenge to manage e.g., grade 3 endome-trioid carcinoma? What is the natural history of ECs withtwo molecular features e.g. MMR-deficient and aberrantp53? How often are these tumors encountered and howshould they be categorized? Is there a surrogate thatcould replace sequencing for POLE? Are favorable out-comes in POLE patients independent of treatment(e.g., can these women be spared adjuvant therapy?)Although there may be many questions to address weanticipate that molecular classification will facilitaterapid progress in research and clinical care as has beenachieved through a subtype specific approach in othertumor sites.In summary, whilst the combination of histomorphol-ogy and clinical factors has proven to be insufficientlyreproducible, prognostic and predictive, two molecularclassifiers based on the TCGA study show great poten-tial as pragmatic and effective tools to stratify patientrisk and subsequent care decisions. Given the high andincreasing incidence of endometrial cancer and the soci-etal cost of over- and under-treatment there is urgentneed for prospective clinical studies to determine howbest to utilize these tools.*For ProMisE; the materials, assay and interpret-ation costs total < $300 USDAbbreviationsEC: Endometrial cancer; TCGA: The cancer genome atlas; ESMO: EuropeanSociety for Medical Oncology; LVSI: Lymph-vascular space invasion; POLEEDM: Polymerase epsilon exonuclease domain mutation; MMR: Mismatchrepair; MMR-D: Mismatch repair deficiencyAcknowledgementsThe authors are part of OVCARE, British Columbia’s gynecologic cancerresearch team and are grateful for the intellectual contributions andencompassing work towards the development and implementation of themolecular classifier. In particular, our lab team of Samuel Leung, MelissaMcConechy, Winnie Yang, Amy Lum, and Janine Senz, mentors andcollaborators Blake Gilks and David Huntsman, and local and internationalcollaborators Janice Kwon, Rob Soslow, Lien Hoang, Martin Kobel, andCheng Han Lee. We are also grateful for supportive funding through theCanadian Institute of Health Research (CIHR New investigator award,McAlpine and CIHR Proof of Principal Phase I grant (201509-PPP-355221-PPP-CAAA-168787)) and the BC Cancer Foundation (Clinical Investigator Award(McAlpine) and the Sarabjit Gill Fund).FundingNo specific funding is associated with this review article. The research fromour center described herein has in part been funded by the sourcesacknowledged above.Availability of data and materialsThe data that support the findings of this study referenced.Authors’ contributionsAT and JM authored this manuscript and continue to work in translationalresearch on endometrial carcinomas. Both authors read and approved thefinal manuscript.Competing interestsUS patent 62192230 for the ProMisE molecular classifier has been filed(pending) by the BC cancer agency.Consent for publicationConsent was given by collaborators in TCGA and TransPORTEC group for thedescription of their material although all material reviewed here is previouslypublished and any information extracted freely accessible to all. No figuresor tables from these or other authors mansucripts are reproduced in thismanuscript thus ‘consent for approval’ does not really apply.Ethics approval and consent to participateThis is a review article, not requiring ethics approval. Endometrial cancerresearch performed at our center for the initiatives described were approvedby the UBC Clinical Research Ethics Board (H09-00939).Author details1Department of Pathology and Laboratory Medicine, University of BritishColumbia and BC Cancer Agency, Vancouver, BC, Canada. 2Department ofGynecology and Obstetrics, Division of Gynecologic Oncology, University ofBritish Columbia, 2775 Laurel St. 6th Floor, Vancouver, BC, CanadaV5Z 1M9.Received: 10 October 2016 Accepted: 22 November 2016References1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin. 2015;65(1):5–29.Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 9 of 122. Society CC. Canadian Cancer. Statistics. 2016;2016(2016):1–142.3. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al.Cancer incidence and mortality worldwide: sources, methods and majorpatterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359–86.4. Rose PG. Endometrial carcinoma. N Engl J Med. 1996;335(9):640–9.5. Creasman WT, Odicino F, Maisonneuve P, Beller U, Benedet JL, Heintz AP,et al. Carcinoma of the corpus uteri. J Epidemiol Biostat. 2001;6(1):47–86.6. Morice P, Leary A, Creutzberg C, Abu-Rustum N, Darai E. Endometrial cancer.Lancet. 2016;387(10023):1094–108.7. Ueda SM, Kapp DS, Cheung MK, Shin JY, Osann K, Husain A, et al. Trends indemographic and clinical characteristics in women diagnosed with corpuscancer and their potential impact on the increasing number of deaths. AmJ Obstet Gynecol. 2008;198(2):218 e1–6.8. Hamilton CA, Cheung MK, Osann K, Chen L, Teng NN, Longacre TA, et al.Uterine papillary serous and clear cell carcinomas predict for poorer survivalcompared to grade 3 endometrioid corpus cancers. Br J Cancer. 2006;94(5):642–6. Pubmed Central PMCID: 2361201.9. del Carmen MG, Birrer M, Schorge JO. Uterine papillary serous cancer: a reviewof the literature. Gynecol Oncol. 2012;127(3):651–61.10. Del Carmen MG, Boruta 2nd DM, Schorge JO. Recurrent endometrial cancer.Clin Obstet Gynecol. 2011;54(2):266–77.11. Jung KW, Won YJ, Kong HJ, Oh CM, Lee DH, Lee JS. Prediction of cancerincidence and mortality in Korea, 2014. Cancer Res Treat. 2014;46(2):124–30.Pubmed Central PMCID: 4022820.12. Lim MC, Moon EK, Shin A, Jung KW, Won YJ, Seo SS, et al. Incidence ofcervical, endometrial, and ovarian cancer in Korea, 1999–2010. J GynecolOncol. 2013;24(4):298–302. Pubmed Central PMCID: 3805909.13. Rahib L, Smith BD, Aizenberg R, Rosenzweig AB, Fleshman JM, Matrisian LM.Projecting cancer incidence and deaths to 2030: the unexpected burden ofthyroid, liver, and pancreas cancers in the United States. Cancer Res. 2014;74(11):2913–21.14. Colombo N, Creutzberg C, Amant F, Bosse T, Gonzalez-Martin A, LedermannJ, et al. ESMO-ESGO-ESTRO consensus conference on endometrial cancer:diagnosis, treatment and follow-up. Int J Gynecol Cancer. 2016;26(1):2–30.Pubmed Central PMCID: 4679344.15. Bestvina CM, Fleming GF. Chemotherapy for endometrial cancer in adjuvantand advanced disease settings. Oncologist. 2016;21(10):1250–9. PubmedCentral PMCID: 5061541.16. Galaal K, Al Moundhri M, Bryant A, Lopes AD, Lawrie TA. Adjuvantchemotherapy for advanced endometrial cancer. Cochrane Database SystRev. 2014;5:CD010681.17. Galaal K, Bryant A, Fisher AD, Al-Khaduri M, Kew F, Lopes AD. Laparoscopyversus laparotomy for the management of early stage endometrial cancer.Cochrane Database Syst Rev. 2012;12(9):CD006655.18. de Boer SM, Powell ME, Mileshkin L, Katsaros D, Bessette P, Haie-Meder C, et al.Toxicity and quality of life after adjuvant chemoradiotherapy versus radiotherapyalone for women with high-risk endometrial cancer (PORTEC-3): an open-label,multicentre, randomised, phase 3 trial. Lancet Oncol. 2016;17(8):1114–26.19. Joly F, McAlpine J, Nout R, Avall-Lundqvist E, Shash E, Friedlander M, et al.Quality of life and patient-reported outcomes in endometrial cancer clinicaltrials: a call for action! Int J Gynecol Cancer. 2014;24(9):1693–9.20. McAlpine JN, Greimel E, Brotto LA, Nout RA, Shash E, Avall-Lundqvist E, et al.Quality of life research in endometrial cancer: what is needed to advanceprogress in this disease site? Methodological considerations from theGynecologic Cancer InterGroup Symptom Benefit Working Group brainstormingsession, Leiden 2012. Int J Gynecol Cancer. 2014;24(9):1686–92.21. Murali R, Soslow RA, Weigelt B. Classification of endometrial carcinoma:more than two types. Lancet Oncol. 2014;15(7):e268–78.22. Bendifallah S, Darai E, Ballester M. Predictive modeling: a new paradigm formanaging endometrial cancer. Ann Surg Oncol. 2016;23(3):975–88.23. Bokhman JV. Two pathogenetic types of endometrial carcinoma. GynecolOncol. 1983;15(1):10–7.24. Gilks CB, Oliva E, Soslow RA. Poor interobserver reproducibility in thediagnosis of high-grade endometrial carcinoma. Am J Surg Pathol. 2013;37(6):874–81.25. Clarke BA, Gilks CB. Endometrial carcinoma: controversies in histopathologicalassessment of grade and tumour cell type. J Clin Pathol. 2010;63(5):410–5.26. Hussein YR, Broaddus R, Weigelt B, Levine DA, Soslow RA. The genomicheterogeneity of FIGO grade 3 endometrioid carcinoma impacts diagnosticaccuracy and reproducibility. Int J Gynecol Pathol. 2016;35(1):16–24.Pubmed Central PMCID: 4934379.27. Han G, Sidhu D, Duggan MA, Arseneau J, Cesari M, Clement PB, et al.Reproducibility of histological cell type in high-grade endometrialcarcinoma. Mod Pathol. 2013;26(12):1594–604.28. Alkushi A, Kobel M, Kalloger SE, Gilks CB. High-grade endometrialcarcinoma: serous and grade 3 endometrioid carcinomas have differentimmunophenotypes and outcomes. Int J Gynecol Pathol. 2010;29(4):343–50.29. Karateke A, Tug N, Cam C, Selcuk S, Asoglu MR, Cakir S. Discrepancy ofpre- and postoperative grades of patients with endometrial carcinoma.Eur J Gynaecol Oncol. 2011;32(3):283–5.30. Sany O, Singh K, Jha S. Correlation between preoperative endometrialsampling and final endometrial cancer histology. Eur J Gynaecol Oncol.2012;33(2):142–4.31. Wang XY, Pan ZM, Chen XD, Lu WG, Xie X. Accuracy of tumor grade bypreoperative curettage and associated clinicopathologic factors in clinicalstage I endometriod adenocarcinoma. Chin Med J. 2009;122(16):1843–6.32. Batista TP, Cavalcanti CL, Tejo AA, Bezerra AL. Accuracy of preoperativeendometrial sampling diagnosis for predicting the final pathology gradingin uterine endometrioid carcinoma. Eur J Surg Oncol. 2016;26.33. Guy MS, Sheeder J, Behbakht K, Wright JD, Guntupalli SR. Comparative outcomesin older and younger women undergoing laparotomy or robotic surgical stagingfor endometrial cancer. Am J Obstet Gynecol. 2016;214(3):350 e1–e10.34. Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Schlaerth JB,Mannel RS, et al. Laparoscopy compared with laparotomy forcomprehensive surgical staging of uterine cancer: GynecologicOncology Group Study LAP2. J Clin Oncol. 2009;27(32):5331–6. PubmedCentral PMCID: 2773219.35. Peters 3rd WA, Andersen WA, Thornton Jr WN, Morley GW. The selectiveuse of vaginal hysterectomy in the management of adenocarcinoma of theendometrium. Am J Obstet Gynecol. 1983;146(3):285–9.36. Cornelison TL, Trimble EL, Kosary CL. SEER data, corpus uteri cancer:treatment trends versus survival for FIGO stage II, 1988–1994. GynecolOncol. 1999;74(3):350–5.37. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. Surgery and postoperative radiotherapy versus surgeryalone for patients with stage-1 endometrial carcinoma: multicentrerandomised trial. PORTEC Study Group. Post Operative Radiation Therapy inEndometrial Carcinoma. Lancet. 2000;355(9213):1404–11.38. Di Cello A, Rania E, Zuccala V, Venturella R, Mocciaro R, Zullo F, et al. Failureto recognize preoperatively high-risk endometrial carcinoma is associatedwith a poor outcome. Eur J Obstet Gynecol Reprod Biol. 2015;194:153–60.39. Zhang C, Hu W, Jia N, Li Q, Hua K, Tao X, et al. Uterine carcinosarcoma andhigh-risk endometrial carcinomas: a clinicopathological comparison. Int JGynecol Cancer. 2015;25(4):629–36.40. Imai K, Kato H, Katayama K, Nakanishi K, Kawano A, Iura A, et al. A preoperativerisk-scoring system to predict lymph node metastasis in endometrial cancerand stratify patients for lymphadenectomy. Gynecol Oncol. 2016;142(2):273–7.41. Colombo N, Preti E, Landoni F, Carinelli S, Colombo A, Marini C, et al.Endometrial cancer: ESMO clinical practice guidelines for diagnosis,treatment and follow-up. Ann Oncol. 2013;24 Suppl 6:vi33–8.42. Mariani A, Dowdy SC, Cliby WA, Gostout BS, Jones MB, Wilson TO, et al.Prospective assessment of lymphatic dissemination in endometrial cancer:a paradigm shift in surgical staging. Gynecol Oncol. 2008;109(1):11–8.Pubmed Central PMCID: 3667391.43. AlHilli MM, Mariani A, Bakkum-Gamez JN, Dowdy SC, Weaver AL,Peethambaram PP, et al. Risk-scoring models for individualized prediction ofoverall survival in low-grade and high-grade endometrial cancer. GynecolOncol. 2014;133(3):485–93.44. Kang S, Lee JM, Lee JK, Kim JW, Cho CH, Kim SM, et al. A Web-basednomogram predicting para-aortic nodal metastasis in incompletely stagedpatients with endometrial cancer: a Korean multicenter study. Int J GynecolCancer. 2014;24(3):513–9.45. Keys HM, Roberts JA, Brunetto VL, Zaino RJ, Spirtos NM, Bloss JD, et al.A phase III trial of surgery with or without adjunctive external pelvicradiation therapy in intermediate risk endometrial adenocarcinoma: agynecologic oncology group study. Gynecol Oncol. 2004;92(3):744–51.46. Barlin JN, Soslow RA, Lutz M, Zhou QC, St Clair CM, Leitao Jr MM, et al.Redefining stage I endometrial cancer: incorporating histology, a binarygrading system, myometrial invasion, and lymph node assessment. Int JGynecol Cancer. 2013;23(9):1620–8. Pubmed Central PMCID: 4405774.47. Kong TW, Chang SJ, Paek J, Lee Y, Chun M, Ryu HS. Risk group criteriafor tailoring adjuvant treatment in patients with endometrial cancer:Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 10 of 12a validation study of the gynecologic oncology group criteria. J GynecolOncol. 2015;26(1):32–9. Pubmed Central PMCID: 4302283.48. Kwon JS, Qiu F, Saskin R, Carey MS. Are uterine risk factors more importantthan nodal status in predicting survival in endometrial cancer? ObstetGynecol. 2009;114(4):736–43.49. Bendifallah S, Canlorbe G, Collinet P, Arsene E, Huguet F, Coutant C, et al.Just how accurate are the major risk stratification systems for early-stageendometrial cancer?. Brit J cancer. 2015;12.50. Hoang LN, McConechy MK, Kobel M, Han G, Rouzbahman M, Davidson B,et al. Histotype-genotype correlation in 36 high-grade endometrialcarcinomas. Am J Surg Pathol. 2013;37(9):1421–32.51. Lax SF, Kurman RJ. A dualistic model for endometrial carcinogenesis basedon immunohistochemical and molecular genetic analyses. Verh Dtsch GesPathol. 1997;81:228–32.52. McConechy MK, Ding J, Cheang MC, Wiegand KC, Senz J, Tone AA, et al.Use of mutation profiles to refine the classification of endometrialcarcinomas. J Pathol. 2012;228(1):20–30. Pubmed Central PMCID: 3939694.53. Mcconechy MK, Ding J, Senz J, Yang W, Melnyk N, Tone AA, et al. Ovarianand endometrial endometrioid carcinomas have distinct CTNNB1 and PTENmutation profiles. Mod Pathol. 2014;27(1):128–34. Pubmed Central PMCID:3915240.54. Hoang LN, Lee YS, Karnezis AN, Tessier-Cloutier B, Almandani N, CoathamM, et al. Immunophenotypic features of dedifferentiated endometrialcarcinoma Insights from BRG1/INI1-deficient tumors. Histopathology. 2016.55. Hoang LN, McConechy MK, Meng B, McIntyre JB, Ewanowich C, GilksCB, et al. Targeted mutation analysis of endometrial clear cellcarcinoma. Histopathology. 2014.56. Alvarez T, Miller E, Duska L, Oliva E. Molecular profile of grade 3endometrioid endometrial carcinoma: is it a type I or type II endometrialcarcinoma? Am J Surg Pathol. 2012;36(5):753–61.57. Santacana M, Maiques O, Valls J, Gatius S, Abo AI, Lopez-Garcia MA,et al. A 9-protein biomarker molecular signature for predictinghistologic type in endometrial carcinoma by immunohistochemistry.Hum Pathol. 2014;45(12):2394–403.58. Coenegrachts L, Garcia-Dios DA, Depreeuw J, Santacana M, Gatius S, ZikanM, et al. Mutation profile and clinical outcome of mixed endometrioid-serous endometrial carcinomas are different from that of pureendometrioid or serous carcinomas. Virchows Archiv. 2015;466(4):415–22.59. Ferguson SE, Olshen AB, Viale A, Barakat RR, Boyd J. Stratification ofintermediate-risk endometrial cancer patients into groups at high risk orlow risk for recurrence based on tumor gene expression profiles. ClinCancer Res. 2005;11(6):2252–7.60. Salvesen HB, Carter SL, Mannelqvist M, Dutt A, Getz G, Stefansson IM, et al.Integrated genomic profiling of endometrial carcinoma associatesaggressive tumors with indicators of PI3 kinase activation. Proc Natl AcadSci U S A. 2009;106(12):4834–9. Pubmed Central PMCID: 2660768.61. Wik E, Trovik J, Kusonmano K, Birkeland E, Raeder MB, Pashtan I, et al.Endometrial Carcinoma Recurrence Score (ECARS) validates to identifyaggressive disease and associates with markers of epithelial-mesenchymaltransition and PI3K alterations. Gynecol Oncol. 2014;134(3):599–606.62. Wik E, Birkeland E, Trovik J, Werner HM, Hoivik EA, Mjos S, et al. Highphospho-Stathmin(Serine38) expression identifies aggressive endometrialcancer and suggests an association with PI3K inhibition. Clin Cancer Res.2013;19(9):2331–41.63. Myers AP. New strategies in endometrial cancer: targeting the PI3K/mTORpathway–the devil is in the details. Clin Cancer Res. 2013;19(19):5264–74.64. Krakstad C, Birkeland E, Seidel D, Kusonmano K, Petersen K, Mjos S, et al.High-throughput mutation profiling of primary and metastatic endometrialcancers identifies KRAS, FGFR2 and PIK3CA to be frequently mutated. PloSone. 2012;7(12):e52795. Pubmed Central PMCID: 3531332.65. Salvesen HB, Haldorsen IS, Trovik J. Markers for individualised therapy inendometrial carcinoma. Lancet Oncol. 2012;13(8):e353–61.66. Iglesias DA, Yates MS, van der Hoeven D, Rodkey TL, Zhang Q, Co NN, et al.Another surprise from Metformin: novel mechanism of action via K-Rasinfluences endometrial cancer response to therapy. Mol Cancer Ther. 2013;12(12):2847–56. Pubmed Central PMCID: 3883498.67. Cancer Genome Atlas Research N, Kandoth C, Schultz N, Cherniack AD,Akbani R, Liu Y, et al. Integrated genomic characterization of endometrialcarcinoma. Nature. 2013;497(7447):67–73. Pubmed Central PMCID: 3704730.68. Meng B, Hoang LN, McIntyre JB, Duggan MA, Nelson GS, Lee CH, et al.POLE exonuclease domain mutation predicts long progression-free survivalin grade 3 endometrioid carcinoma of the endometrium. Gynecol Oncol.2014;134(1):15–9.69. Stelloo E, Bosse T, Nout RA, MacKay HJ, Church DN, Nijman HW, et al.Refining prognosis and identifying targetable pathways for high-riskendometrial cancer; a TransPORTEC initiative. Mod Pathol. 2015.70. Billingsley CC, Cohn DE, Mutch DG, Stephens JA, Suarez AA, Goodfellow PJ.Polymerase varepsilon (POLE) mutations in endometrial cancer: Clinicaloutcomes and implications for Lynch syndrome testing. Cancer. 2015;121(3):386–94. Pubmed Central PMCID: 4304930.71. Stelloo E, Nout RA, Osse EM, Jurgenliemk-Schulz IJ, Jobsen JJ, Lutgens LC, et al.Improved risk assessment by integrating molecular and clinicopathologicalfactors in early-stage endometrial cancer - combined analysis of PORTECcohorts. Clin Cancer Res. 2016;22.72. Church DN, Stelloo E, Nout RA, Valtcheva N, Depreeuw J, ter Haar N, et al.Prognostic significance of POLE proofreading mutations in endometrialcancer. J Natl Cancer Inst. 2015;107(1):402.73. Talhouk A, McConechy MK, Leung S, Li-Chang HH, Kwon JS, Melnyk N, et al.A clinically applicable molecular-based classification for endometrial cancers.Br J Cancer. 2015;113(2):299–310. Pubmed Central PMCID: 4506381.74. McConechy MK, Talhouk A, Leung S, Chiu DS, Yang W, Senz J, et al.Endometrial carcinomas with POLE exonuclease domain mutations have afavorable prognosis. Clin Cancer Res. 2016;13.75. Cancer Genome Atlas Research N, Weinstein JN, Collisson EA, Mills GB, ShawKR, Ozenberger BA, et al. The cancer genome atlas pan-cancer analysisproject. Nat Genet. 2013;45(10):1113–20. Pubmed Central PMCID: 3919969.76. Umar A, Boland CR, Terdiman JP, Syngal S, de la Chapelle A, Ruschoff J,et al. Revised Bethesda guidelines for hereditary nonpolyposis colorectalcancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst.2004;96(4):261–8. Pubmed Central PMCID: 2933058.77. Talhouk A, McConechy M, Leung S, Yang W, Lum A, Senz J, et al.Confirmation of ProMisE: a simple genomics-based clinical classifier forendometrial cancer. Cancer. 2016; Accepted, In Press.78. In: Micheel CM, Nass SJ, Omenn GS, editors. Evolution of translational omics:Lessons learned and the path forward. Washington (DC): NationalAcademies Press; 2012.79. Talhouk A, McConechy MK, Leung S, Yang W, Senz J, et al. A clinicallyapplicable molecular-based classification system for endometrial cancers.J Clin Oncol. 2016; ASCO poster presention: Abstract No: 5518.80. Talhouk A, Hoang LN, McConechy MK, Nakonechny Q, Leo J, Cheng A, et al.Molecular classification of endometrial carcinoma on diagnostic specimensis highly concordant with final hysterectomy: earlier prognostic informationto guide treatment. Gynecol Oncol. 2016;143(1):46–53.81. Stelloo E, Nout RA, Naves LC, Ter Haar NT, Creutzberg CL, Smit VT, et al.High concordance of molecular tumor alterations between pre-operativecurettage and hysterectomy specimens in patients with endometrialcarcinoma. Gynecol Oncol. 2014;133(2):197–204.82. Visser NC, Bulten J, van der Wurff AA, Boss EA, Bronkhorst CM, Feijen HW, et al.PIpelle Prospective ENDOmetrial carcinoma (PIPENDO) study, pre-operativerecognition of high risk endometrial carcinoma: a multicentre prospectivecohort study. BMC Cancer. 2015;15:487. Pubmed Central PMCID: 4485884.83. McConechy MK, Talhouk A, Li-Chang HH, Leung S, Huntsman DG, Gilks CB,et al. Detection of DNA mismatch repair (MMR) deficiencies byimmunohistochemistry can effectively diagnose the microsatellite instability(MSI) phenotype in endometrial carcinomas. Gynecol Oncol. 2015;28.84. Goodfellow PJ, Billingsley CC, Lankes HA, Ali S, Cohn DE, Broaddus RJ, et al.Combined microsatellite instability, MLH1 methylation analysis, andimmunohistochemistry for lynch syndrome screening in endometrialcancers from GOG210: An NRG oncology and gynecologic oncology groupstudy. J Clin Oncol Off J Am Soc Clin Oncol. 2015;33(36):4301–8. PubmedCentral PMCID: 4678181.85. Hall G, Clarkson A, Shi A, Langford E, Leung H, Eckstein RP, et al.Immunohistochemistry for PMS2 and MSH6 alone can replace a fourantibody panel for mismatch repair deficiency screening in colorectaladenocarcinoma. Pathology. 2010;42(5):409–13.86. Burgart LJ. Testing for defective DNA mismatch repair in colorectalcarcinoma: a practical guide. Arch Pathol Lab Med. 2005;129(11):1385–9.87. Yamamoto H, Imai K. Microsatellite instability: an update. Arch Toxicol. 2015;89(6):899–921.88. Rabban JT, Calkins SM, Karnezis AN, Grenert JP, Blanco A, Crawford B, et al.Association of tumor morphology with mismatch-repair protein status inolder endometrial cancer patients: implications for universal versusTalhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 11 of 12selective screening strategies for Lynch syndrome. Am J Surg Pathol.2014;38(6):793–800.89. Garg K, Soslow RA. Lynch syndrome (hereditary non-polyposiscolorectal cancer) and endometrial carcinoma. J Clin Pathol. 2009;62(8):679–84.90. Bonadona V, Bonaiti B, Olschwang S, Grandjouan S, Huiart L, Longy M, et al.Cancer risks associated with germline mutations in MLH1, MSH2, and MSH6genes in Lynch syndrome. Jama. 2011;305(22):2304–10.91. Hendriks YM, Wagner A, Morreau H, Menko F, Stormorken A, QuehenbergerF, et al. Cancer risk in hereditary nonpolyposis colorectal cancer due toMSH6 mutations: impact on counseling and surveillance. Gastroenterology.2004;127(1):17–25.92. Carayol J, Bonaiti-Pellie C. Estimating penetrance from family data using aretrospective likelihood when ascertainment depends on genotype and ageof onset. Genet Epidemiol. 2004;27(2):109–17.93. Bonaiti B, Bonadona V, Perdry H, Andrieu N, Bonaiti-Pellie C. Estimatingpenetrance from multiple case families with predisposing mutations:extension of the ‘genotype-restricted likelihood’ (GRL) method. Eur J HumGenet. 2011;19(2):173–9. Pubmed Central PMCID: 3025788.94. McMeekin DS, Tritchler DL, Cohn DE, Mutch DG, Lankes HA, Geller MA, et al.Clinicopathologic significance of mismatch repair defects in endometrialcancer: an NRG oncology/gynecologic oncology group study. J Clin Oncol.2016;34(25):3062–8. Pubmed Central PMCID: 5012715.95. Shikama A, Minaguchi T, Matsumoto K, Akiyama-Abe A, Nakamura Y,Michikami H, et al. Clinicopathologic implications of DNA mismatch repairstatus in endometrial carcinomas. Gynecol Oncol. 2016;140(2):226–33.96. Bakhsh S, Kinloch M, Hoang LN, Soslow R, Kobel M, Lee CH, et al. Histopathologicalfeatures of endometrial carcinomas associated with POLE mutations: implicationsfor decisions about adjuvant therapy. Histopathology. 2015.97. Bellone S, Centritto F, Black J, Schwab C, English D, Cocco E, et al.Polymerase epsilon (POLE) ultra-mutated tumors induce robust tumor-specific CD4+ T cell responses in endometrial cancer patients. GynecolOncol. 2015;138(1):11–7. Pubmed Central PMCID: 4469551.98. Hussein YR, Weigelt B, Levine DA, Schoolmeester JK, Dao LN, Balzer BL, et al.Clinicopathological analysis of endometrial carcinomas harboring somaticPOLE exonuclease domain mutations. Mod Pathol. 2015;28(4):505–14.99. Howitt BE, Shukla SA, Sholl LM, Ritterhouse LL, Watkins JC, Rodig S, et al.Association of polymerase e-mutated and microsatellite-instable endometrialcancers with neoantigen load, number of tumor-infiltrating lymphocytes, andexpression of PD-1 and PD-L1. JAMA Oncol. 2015;1(9):1319–23.100. Gargiulo P, Della Pepa C, Berardi S, Califano D, Scala S, Buonaguro L, et al.Tumor genotype and immune microenvironment in POLE-ultramutated andMSI-hypermutated endometrial cancers: new candidates for checkpointblockade immunotherapy? Cancer Treat Rev. 2016;48:61–8.101. van Gool IC, Bosse T, Church DN. POLE proofreading mutation, immuneresponse and prognosis in endometrial cancer. Oncoimmunology. 2016;5(3):e1072675. Pubmed Central PMCID: 4839358.102. Santin AD, Bellone S, Buza N, Choi J, Schwartz PE, Schlessinger J, et al.Regression of chemotherapy-resistant Polymerase epsilon (POLE) ultra-mutated and MSH6 hyper-mutated endometrial tumors with nivolumab.Clin Cancer Res. 2016.103. Mehnert JM, Panda A, Zhong H, Hirshfield K, Damare S, Lane K, et al. Immuneactivation and response to pembrolizumab in POLE-mutant endometrialcancer. J Clin Invest. 2016;126(6):2334–40. Pubmed Central PMCID: 4887167.104. Schultheis AM, Martelotto LG, De Filippo MR, Piscuglio S, Ng CK, Hussein YR,et al. TP53 mutational spectrum in endometrioid and serous endometrialcancers. Int J Gynecol Pathol. 2016;35(4):289–300.105. Hansen JM, Baggerly KA, Wang Y, Wu S, Previs RA, Zand B, et al.Homologous recombination deficiency in endometrioid uterine cancer: anunrecognized phenomenon. Gynecol Oncol. 2015;137(1):21.106. Westin SN, Ju Z, Broaddus RR, Krakstad C, Li J, Pal N, et al. PTEN loss is acontext-dependent outcome determinant in obese and non-obeseendometrioid endometrial cancer patients. Mol Oncol. 2015;9(8):1694–703.Pubmed Central PMCID: 4584169.107. Roque DR, Makowski L, Chen TH, Rashid N, Hayes DN, Bae-Jump V.Association between differential gene expression and body mass indexamong endometrial cancers from the cancer genome atlas project. GynecolOncol. 2016;142(2):317–22. Pubmed Central PMCID: 4961559.108. Berg A, Hoivik EA, Mjos S, Holst F, Werner HM, Tangen IL, et al. Molecularprofiling of endometrial carcinoma precursor, primary and metastaticlesions suggests different targets for treatment in obese compared tonon-obese patients. Oncotarget. 2014.109. Hoang LN, Han G, McConechy M, Lau S, Chow C, Gilks CB, et al.Immunohistochemical characterization of prototypical endometrial clear cellcarcinoma–diagnostic utility of HNF-1beta and oestrogen receptor.Histopathology. 2014;64(4):585–96.110. Coatham M, Li X, Karnezis AN, Hoang LN, Tessier-Cloutier B, Meng B, et al.Concurrent ARID1A and ARID1B inactivation in endometrial and ovariandedifferentiated carcinomas. Mod Pathol. 2016.111. Karnezis AN, Hoang LN, Coatham M, Ravn S, Almadani N, Tessier-Cloutier B,et al. Loss of switch/sucrose non-fermenting complex protein expression isassociated with dedifferentiation in endometrial carcinomas. Mod Pathol.2016;29(3):302–14. Pubmed Central PMCID: 4980656.112. Kobel M, Meng B, Hoang LN, Almadani N, Li X, Soslow RA, et al. Molecularanalysis of mixed endometrial carcinomas shows clonality in most cases.Am J Surg Pathol. 2016;40(2):166–80.113. Gibson WJ, Hoivik EA, Halle MK, Taylor-Weiner A, Cherniack AD, Berg A,et al. The genomic landscape and evolution of endometrial carcinomaprogression and abdominopelvic metastasis. Nat Genet. 2016;48(8):848–55.Pubmed Central PMCID: 4963271.114. Wang Y, McConechy M, Gilks B, Huntsman D, Shah SP, McAlpine J.Genome-wide copy number analysis and mutational profiling providesevidence of intratumoral heterogeneity in endometrial cancers. Int J GyneCancer. 2014;24(Suppl 49).115. Kobel M, Reuss A, du Bois A, Kommoss S, Kommoss F, Gao D, et al. Thebiological and clinical value of p53 expression in pelvic high-grade serouscarcinomas. J Pathol. 2010;222(2):191–8.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Talhouk and McAlpine Gynecologic Oncology Research and Practice  (2016) 3:14 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items