Testing for POLE mutations, a molecular method of categorising endometrial cancers, provides more accurate information than conventional parameters and may be useful for doctors involved in managing gynaecological cancers.
Limitations of traditional classification of high-grade endometrial carcinomas
Classification of endometrial cancers (ECs) by histology alone is poorly reproducible and imperfectly reflects tumour biology, particularly for high-grade (Grade 3) EC1,2. This, in turn, can lead to imprecise estimation of the recurrence risk and may lead to over- or under-treatment of patients.
TCGA molecular classification of endometrial carcinomas
In 2013, analysis of The Cancer Genome Atlas (TCGA) identified four molecular subclasses of EC, with distinct prognostic differences. The categories were POLE ultra-mutated, MSI (microsatellite instability) hyper-mutated, copy-number low, and copy-number high.
POLE ultra-mutated cancers harbour mutations in the exonuclease domain of the polymerase epsilon (POLE) gene and patients can expect an excellent prognosis. Microsatellite unstable EC (MSI, driven by mismatch repair deficiency) and copy-number low EC (also termed no specific molecular profile [NSMP]) have an intermediate outcome. Tumours in the copy-number high subclass typically harbour TP53 mutations and are the most aggressive cancers, with unfavourable prognosis.5
Whilst significant, the findings of the TCGA were difficult to replicate in routine clinical practice, due to cost and the requirement for fresh-frozen tumour samples.
The ProMisE classifier
ProMisE (Proactive Molecular Risk Classifier for Endometrial Cancer) is a classification system based on TCGA that utilises pragmatic methods to identify these genomic subgroups.
Classification is feasible on formalin-fixed, paraffin-embedded (FFPE) material, and assigns patients with EC to one of four groups, based on a combination of immunohistochemistry for MMR proteins (encoded by the genes MLH1, PMS2, MSH2 and MSH6), sequencing for POLE mutation and immunohistochemistry for the p53 protein. Utilising the classifier, endometrial carcinomas are classified into:
p53 wild type (p53-wt); also termed NSMP (no specific molecular profile)
p53 mutant (p53-abn)
Refer to the Endometrial Cancer ProMisE Classifier page for more details by clicking here.
ProMisE classifier – the evidence
The prognostic relevance of this categorisation has been confirmed by analysis of large cohorts with mature follow-up data by the Vancouver and PORTEC groups.4, 5, 27, 28
In the Vancouver cohort, POLE-mutated EC had an adjusted hazard ratio (HR) of 0.15 (p=0.016) for recurrence-free survival and p53-abn EC had an adjusted HR of 2.19 (p=0.016), when both were compared with p53-wt.4
In the large cohort of the PORTEC group, patients with POLE-mutated EC had no locoregional recurrences and p53-abn ECs had an HR of 6.69 (p<0.001) for locoregional recurrence, when both were compared with p53-wt.24
For both groups, an integrated molecular model provided better risk stratification when compared to a model using established clinic-pathological risk factors alone. The data were particularly strong for women with ECs that would currently be categorised as intermediate-high-risk.27, 28
Clinical significance of POLE-mutations
POLE mutations occur in 7-12% of ECs and often occur at a relatively young age.3, 28 They are usually high-grade tumours with adverse histopathologic features that would be typically treated with a high-risk protocol. Despite their aggressive appearance, these tumours have an excellent prognosis. The identification of POLE mutations has the potential to reduce the risks of overtreatment in relatively young patients. In addition, POLE-mutated ECs show both a strong anti-tumour immune response and up-regulation of markers involved in immune suppression, including programmed death 1 (PD-1) and programmed death ligand 1 (PD-L1).12, 13, 23 Therefore, although POLE-mutated ECs rarely recur, patients with recurring or metastatic POLE-mutated ECs may be good candidates for immune checkpoint inhibition.37, 38
In a small proportion of cases, POLE mutations may be inherited (germline), rather than somatic. Germline POLE mutations are associated with a hereditary cancer syndrome, with predisposition to colorectal cancer, along with endometrial and other cancer types.47 If germline origin is suspected for a POLE mutation, referral to a familial cancer or clinical genetics service should be considered.
Histologic characteristics of POLE-mutated ECs
POLE-mutated ECs are often high-grade endometrioid EC with a superficial broad front invasion pattern.14, 16 Tumour giant cells are present in 30% and prominent tumour-infiltrating lymphocytes (TILs) are frequently identified.12-15 Morphologically ambiguous ECs are more likely to be POLE mutated.16 As histological features alone are not conclusive, sequencing of POLE remains the gold standard for identifying cancers with POLE mutations.
A small proportion of ECs cannot be placed in a single ProMisE category. Such cancers have been called ‘multiple classifiers’.8 This raises the question as to which category will reflect the observed natural history of such cancers. There is insufficient data to assign these cancers to one of the four categories or to create an additional category. It is noteworthy that tumours that are both POLE-mutated (favourable prognosis) and p53-abn (unfavourable) have clinical outcomes dictated by the favourable POLE status.46
ProMisE and treatment decisions
Molecular classification of ECs is reproducible and enables more precise management of a patient’s EC.
Pre-operatively, the ProMisE classification may enable earlier and more informed decision-making and surgical planning.7 Surgical staging and lymph node dissection may be omitted for ECs with POLE mutations. Molecular classification may also become relevant for patients who are poor surgical candidates (for example, young women who desire fertility; elderly and high surgical risk patients).
Post-operatively, the ProMisE classification is likely to be integrated into decisions about adjuvant treatment. The PORTEC-4a trial (currently recruiting) is the first trial that uses molecular features to study the effects of differential adjuvant treatments on EC outcomes.31 PORTEC-4a will compare standard adjuvant treatment for high-intermediate-risk (HIR) EC patients (vaginal brachytherapy) with different adjuvant treatments based on the integrated molecular profile.
For follow-up, patients with p53-abn ECs may need closer monitoring than those with POLE-mutated tumours.
For women with recurrent EC, ProMisE could inform targeted treatment and identify those most likely to derive benefit. Clinical trials can be directed at treatments for specific molecular categories of patients, for example, immuno-therapy in patients with progressive MMR deficient EC.33, 34
The molecular classification of EC potentially provides more precise management of endometrial cancer. A key advantage of the molecular classification is that it is robust and reproducible. Nonetheless, the relative significance of clinical, histological and molecular parameters in guiding therapy need to be further refined through clinical trials.
The cost for next-generation sequencing analysis of markers for EC, including POLE mutations, is presented here. Mismatch repair and p53 IHC are required for the ProMisE genomic classifier. Please contact your local Sonic Healthcare pathology practice to arrange. Additional charges may apply. Medicare rebates are not available.
The turnaround time is up to 10 business days, and results can be accessed electronically via Sonic Dx, by fax, or by phone.