Molecular subtypes of clear cell carcinoma of the endometrium: Opportunities for prognostic and predictive stratification

https://doi.org/10.1016/j.ygyno.2020.04.043Get rights and content

Highlights

  • There is molecular heterogeneity within clear cell carcinoma (CCC) of the endometrium.

  • P53wt CCC of endometrium show aggressive features and are distinct from other p53wt endometrial cancers.

  • Molecular classification of CCC of endometrium elicits prognostic parameters that are not apparent with histology alone.

Abstract

Objective

Our aim was to characterize the pathological, molecular and clinical outcomes of clear cell carcinoma of the endometrium (CCC).

Methods

CCC underwent ProMisE (Proactive Molecular Risk Classifier for Endometrial Cancer) classification identifying four molecular subtypes: i) ‘POLEmut’ for ECs with pathogenic POLE mutations, ii) ‘MMRd’, if there is loss of mismatch repair proteins by immunohistochemistry (IHC), iii) ‘p53wt’ or iv) ‘p53abn’ based on p53 IHC staining. Clinicopathologic parameters, immune markers (CD3, CD8, CD79a, CD138, PD-1), ER, L1CAM, and outcomes were assessed.

Results

52 CCCs were classified, including 1 (2%) POLEmut, 5 (10%) MMRd, 28 (54%) p53wt and 18 (35%) p53abn. Women with p53abn and p53wt CCCs were older than women with MMRd and POLEmut subtypes. p53wt CCC were distinct from typical p53wt endometrioid carcinomas; more likely to arise in older, thinner women, with advanced stage disease, LVSI and lymph node involvement, and a higher proportion ER negative, L1CAM overexpressing tumors with markedly worse outcomes. High levels of immune infiltrates (TILhigh) were observed in 75% of the CCC cohort. L1CAM overexpression was highest within p53abn and p53wt subtypes of CCC.

Conclusion

CCC is a heterogeneous disease encompassing all four molecular subtypes and a wide range of clinical outcomes. Outcomes of patients with POLEmut, MMRd and p53abn CCC are not distinguishable from those of other patients with these respective subtypes of EC; p53wt CCC, however, differ from endometrioid p53wt EC in clinical, pathological, molecular features and outcomes. Thus, p53wt CCC of endometrium appear to be a distinct clinicopathological entity within the larger group of p53wt ECs.

Introduction

Endometrial carcinoma (EC) is the most common gynecologic malignancy in North America [1]. Clear cell carcinoma (CCC) of the endometrium accounts for <6% of all endometrial carcinomas and has historically been considered an aggressive ‘Type II’ endometrial cancer, with a propensity for extrauterine spread and poor prognosis [[2], [3], [4]]. Surgical management and adjuvant therapy for CCC usually follows the same treatment algorithms of uterine serous carcinoma although the rationale for this strategy has been questioned in recent years [4].

Challenges of histomorphological categorization of endometrial carcinoma, particularly in high grade tumors are well known [5,6] with inter-observer disagreement in up to one-third of cases [6]. This lack of reproducibility is a major barrier to risk stratification as treatment recommendations are primarily based on these unreliable pathological parameters (histotype, grade), and variation in treatment within and across centers can be dramatic [4]. A move to more reproducible methods of classification through characterization of molecular features has proven to be feasible, consistent, and provides biologically relevant information for patients and clinicians [[7], [8], [9], [10], [11]].

The Cancer Genome Atlas (TCGA) was the first study to comprehensively classify endometrial cancers into genomic subgroups. Pragmatic classification systems using low cost clinically applicable tools on standard formalin fixed paraffin-embedded (FFPE) material e.g. the ProMisE classifier, followed [8,9] identifying 4 molecular subtypes with distinct prognostic outcomes. Key distinguishing parameters include i) identifying pathogenic mutations in the exonuclease domain of DNA polymerase epsilon involved in DNA proofreading/repair (POLEmut), ii) the presence or absence of proteins associated with DNA mismatch repair where ‘loss’ identifies MMR deficiency (MMRd), and iii) p53 status where p53 wildtype has normal expression on IHC (‘p53wt’, also termed ‘no specific molecular profile’ or ‘NSMP’ by the PORTEC/Leiden investigators) and abnormal expression includes both complete loss of detectable p53 and/or overexpression (‘p53abn’). The majority of work on the molecular classification of endometrial cancers has been in endometrioid and serous histotypes and the TCGA excluded other histologies entirely. The PORTEC 1 and 2 series, Vancouver discovery and confirmation cohorts, and German validation series classified only 18, 5 and 6 CCC in total respectively and demonstrated that CCCs can show features of different molecular subtypes [8,[10], [11], [12], [13]]. Additional series that have used selected mutational profiling and/or IHC independent of TCGA-based classification systems also support the molecular diversity of CCC [[13], [14], [15], [16]]. Yet treatment algorithms direct the same aggressive surgery and adjuvant therapy for all CCC. It is time to ask whether the clinical behavior of CCC is also diverse and whether treatment can be tailored according to molecular subtype.

Herein, we sought to characterize the molecular diversity of endometrial CCC, assessing prognostic and predictive biomarkers within the framework of the ProMisE molecular classifier.

Section snippets

Patient cohort, clinical data and outcomes

Institutional Review Board (IRB) approval for clinical, molecular and outcome characterization of CCC of the endometrium was obtained. Search of archival and tissue bank material for ECs diagnosed between 1986 and 2016 at the University of British Columbia revealed 92 CCC of which 16 were excluded following pathology review (recategorized), 24 for insufficient tissue or inadequate (2-year minimum) clinical follow-up. Demographic, clinicopathological, treatment, and outcome data

Clinicopathologic variables and ProMisE molecular subtype distribution

There were 52 cases of CCC of endometrium with tissues available for DNA extraction and IHC as well as clinicopathological data with a minimum 2-year follow-up interval (Table 1). A total of 11 women experienced recurrence of their disease, with 13 disease specific deaths, and 19 deaths recorded in total. The median follow-up time for censored patients was 3.4 years (IQR 0.6–4.5). Molecular subtype distribution included 5 cases (9.6%) of MMRd, 1(1.9%) POLEmut, 28 (53.8%) p53wt/NSMP, and 18

Discussion

Historic dualistic models of endometrial carcinoma categorize clear cell carcinomas as ‘Type II’ pathogenetic type with no identified endocrine or metabolic influences on its development and with an aggressive clinical course. Clinical management of CCC has followed recommendations for the more common serous carcinoma, the prototypical ‘Type II’ tumor; however, more recently the mutational diversity within CCCs has been appreciated and this strategy questioned. CCC has been understudied,

CRediT authorship contribution statement

Rachel Soyoun Kim:Methodology, Formal analysis, Data curation, Writing - original draft, Writing - review & editing, Visualization.Basile Tessier Cloutier:Data curation.Samuel Leung:Formal analysis, Software.Dawn Cochrane:Resources, Data curation, Investigation.Heidi Britton:Resources, Data curation, Investigation.Annick Pina:Resources, Data curation, Investigation.Claudine Storness-Bliss:Resources, Data curation, Investigation.David Farnell:Resources, Data curation, Investigation.Leo Huang:

Acknowledgement

The authors would like to thank Simon Cheung and Christine Chow for their assistance with retrieving archival cases and immunohistochemistry. We have received funding from the Michael Smith Foundation for Health Research, the Canadian Institutes of Health Research and the BC Cancer Foundation Clinician Scientist Award pertaining to this work.

Declaration of competing interest

The authors have no conflicts of interest to declare.

References (38)

  • Canadian Cancer Society

    Canadian Cancer Society’s Advisory Committee on Cancer Statistics

    (2015)
  • P.B. Clement et al.

    Non-endometrioid carcinomas of the uterine corpus: a review of their pathology with emphasis on recent advances and problematic aspects

    Adv. Anat. Pathol.

    (2004)
  • C.A. Hamilton et al.

    Uterine papillary serous and clear cell carcinomas predict for poorer survival compared to grade 3 endometrioid corpus cancers

    Br. J. Cancer

    (2006)
  • C.B. Gilks et al.

    Poor interobserver reproducibility in the diagnosis of high-grade endometrial carcinoma

    Am. J. Surg. Pathol.

    (2013)
  • G. Getz et al.

    Integrated genomic characterization of endometrial carcinoma

    Nature.

    (2013)
  • A. Talhouk et al.

    Confirmation of ProMisE: a simple, genomics-based clinical classifier for endometrial cancer

    Cancer.

    (2017)
  • A. Talhouk et al.

    New classification of endometrial cancers: the development and potential applications of genomic-based classification in research and clinical care

    Gynecol Oncol Res Pract.

    (2016)
  • E. Stelloo et al.

    Improved risk assessment by integrating molecular and clinicopathological factors in early-stage endometrial cancer-combined analysis of the PORTEC cohorts

    Clin. Cancer Res.

    (2016)
  • A. Talhouk et al.

    A clinically applicable molecular-based classification for endometrial cancers

    Br. J. Cancer

    (2015)
  • Cited by (79)

    View all citing articles on Scopus
    View full text