Molecular Classification for Endometrial Cancer Revisited

Alexander M. Castellino, PhD

July 30, 2015

Researchers have described a new method for identifying the four molecular subtypes of endometrial cancer. They say their method can be used in routine clinical practice and could alter   management of some patients.

The four subtypes of endometrial cancer are associated with different clinical outcomes, and so identifying them has potential to provide prognostic and predictive information for physicians   and their patients.

They were identified in a molecular classification based on the Cancer Genome Atlas (TCGA), but the cost and   complexity of using whole-genome sequencing makes the approach unsuitable for wider applicability.

Now, a group from the University of British Columbia has bridged the gap by showing that this molecular classification is possible in routine clinical practice on formalin-fixed paraffin-  embedded samples by applying "simple, lower-cost, molecular-based classification methodologies that can recover the TCGA subtypes." The study was published online June 30 in the British Journal of Cancer.

"Molecular features can more consistently classify endometrial cancers," corresponding author Jessica McAlpine, MD, associate professor in the Department of Gynecology at the University of   British Columbia, Vancouver, Canada, told Medscape Medical News.

"The molecular classification proposed by the TCGA cannot be sustained in clinical practice because it is costly and complex. The surrogate tests we have proposed have recovered the same   molecular subgroups," she added.

"The survival curves [are] reproduced with assays that could be used in routine clinical practice," Dr McAlpine and her colleagues write.

Recovering the TCGA Subtypes

The four molecular subtypes of EC proposed by the TCGA are as follows:

  1. POLE mutations and their ultramutated phenotype associated with highly favorable clinical outcomes

  2. Microsatellite instability (MSI)

  3. Copy-number low (CN low)

  4. Copy-number high (CN high), consisting mainly of high-grade serous cancers

By analyzing the molecular subgroups, the Vancouver group was able to identify molecular surrogates corresponding to each of the subgroups.

However, the surrogates, or this new classifier tool, had to successfully classify all patients in the TCGA cohort and had to minimize false negatives in the CN-high poor prognosticator   group.

The researchers identified the POLE phenotype on the basis of a POLE mutation and also included PTEN mutations in an initial classification, because it occurred in all   "ultramutated" phenotypes. Dr McAlpine indicated that currently, there is no surrogate for POLE mutation but that targeted sequencing for the common mutations in this gene could be used   rather than whole-genome or panel testing, which is more costly.

She was hopeful that in the next 2 to 5 years, POLE testing will be available at any center.

The MSI assay was substituted by immunohistochemistry (IHC) on four mismatch repair (MMR) proteins (MLH1, MSH2, MSH6, and PMS2). "MMR IHC testing is cost-effective, practical, and readily   available," Dr McAlpine told Medscape Medical News.

"Testing for MMR IHC has the additional benefit of identifying women who might benefit from genetic testing for an inherited condition called Lynch syndrome. This would have implications to the   individual and to her family members," she added.

The CN-high surrogate was identified by mining the TCGA microarray data. CN status was defined by three genetic loci: FGFR (4p16.3), SOX17 (8q11.23), and MYC (8q24.12).   These loci were able to identify all the CN-high cases.

In addition, aberrant/abnormal (abn) p53 by genetic testing or IHC for complete loss or overexpressing (2+) was able to separate CN-high (p53 abn) from CN-low (normal p53) subtypes.

Reproducing the Data

Using several models, which consisted of varying the combinations of the molecular features described, allowed the researchers to reproduce the four genomic subgroups and survival curves for   the TCGA cohort.

POLE mutation, MMR IHC abn, p53 wild type, and p53 abn were significant in replicating the four TCGA molecular subtypes.

Applying this same rigor to 152 patients with EC identified from the Vancouver General Hospital cases banked in the OVCARE Tissue Bank Repository, Dr McAlpine and colleagues replicated the   subgroups and the survival curves for the test cohort.

Significantly, Dr McAlpine and colleagues showed that although current clinical risk stratification based on pathologic parameters was also associated with clinical outcomes, such as overall   survival and relapse-free survival, combining pathologic parameters and molecular classification made it possible to better discriminate EC outcomes and to do so more confidently.

The favored model for molecular classification first looks at MMR IHC. Tumor samples that do not show MMR IHC abn are then subject to POLE mutation. POLE WT samples are further tested   for p53: p53 WT is scored on the basis of p53 IHC (1+), and p53 abn is scored on the basis of p53 IHC (0 or 2+).

Significance of Molecular Classification in EC

The data presented in this analysis show that the clinicopathologic risk groups are not equivalent to the molecular subgroups identified. Approximately 50% of the POLE subgroup was   identified as high risk, as determined on the basis of grade, stage, and/or histotype; that subgroup would have received additional therapy. This may not have been needed, given their highly   favorable outcomes, Dr McAlpine noted.

So how could the molecular features of EC guide clinical practice? "In endometrial cancer, treatment is based on clinical risk features, such as grade and histology, which are only available   after hysterectomy," Dr McAlpine told Medscape Medical News.

"This is late in the decision-making process," she added. These molecular tests can be performed on endometrial biopsy specimens obtained in doctors' offices; such tests are almost always the   first step in diagnosis in these women. A young patient with a POLE mutation may not need surgery right away and may be able to be successfully managed with hormonal therapy, whereas a   woman identified as having abnormal mismatch repair and who tests positive for Lynch syndrome would be encouraged to have definitive surgery, Dr McAlpine explained.

Dr McAlpine and colleagues emphasize the importance of molecular testing in the clinical management of EC. "If we can continue to demonstrate equivalence of a molecular classification system in   diagnostic endometrial samples and prognostic significance of a classifier in new and larger cohorts, as we are doing, then women and their physicians could have valuable information that would   help them guide decision making at the earliest time point in their cancer journey (eg, at diagnosis)," the authors write.

"Decisions could be made before surgical staging regarding the urgency and extent of surgery, anticipated adjuvant therapy and follow-up plans. This information would be particularly helpful in   guiding young women, with 14% of endometrial cancers arising in women <50 years of age and 5% in women younger than 40," they add.

Next Steps

Dr McAlpine indicated that the group has completed an analysis of 400 additional samples that was based on the Institute of Medicine guidelines to corroborate the data in an independent sample   set.

In addition, they are testing a cohort for endometrial biopsy specimens vs samples obtained from hysterectomies to determine the level of concordance between the two cohorts.

"If endometrial biopsies reflect disease state and are consistent with information available from hysterectomies, physicians and patients will be in a position to make early treatment decisions   more confidently," Dr McAlpine told Medscape Medical News.

In addition, the robustness of the approach is being validated at three different centers, she indicated.

As far as bringing this approach into clinical practice, Dr McAlpine is optimistic that within the next 5 years, significant strides will be made in advancing molecular testing for subtypes   into clinical practice.

"Our goal is to improve upon the current system of clinicopathologic risk group stratification that is based on stage and the irreproducible variables of grade and histotype assignment, and is   not highly predictive of outcomes," the authors write.

"These [molecular] subgroups are associated with clinical outcomes, and identify women who may have a risk of recurrence of their EC that is very different than what is designated by   traditional clinical risk group assessment," Dr McAlpine and her colleagues state.

"Molecular classification in ECs would also allow stratification of cases for clinical trials and assessment of treatment efficacy within specific molecular subgroups. This has been a game-  changing approach in ovarian cancers and has the potential to greatly advance progress in endometrial cancer research," they conclude.

This study was supported by the Sarabjit Gill Fund and the BC Cancer Foundation. The authors have disclosed no relevant financial relationships.

Br J Cancer. Published online June 30, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.