Guofeng Gao, M.D., Ph.D., Pathology Resident
Jeffrey Gregg, MD, Senior Director of Clinical Pathology, Director of Molecular Diagnostics
Anthony Karnezis, M.D., Ph.D., Associate Professor, Anatomic Gynecology Surgical Pathology

INTRODUCTION

Endometrial cancer (EC) is the most common gynecologic cancer in the U.S., and it is the 4th most common cancer in women. There are ~61,000 cases annually which account for approximately 7% of all cancers in women (Siegel et al., 2019). Low grade endometrioid carcinoma is the most common endometrial cancer, and 2/3 of these cases present at early stage and have generally good survival (80% overall 5 year survival); however, a subset of these tumors have poor outcomes. An essential question is how to identify tumors at high risk of recurrence because it has important implications for surgical staging and/or adjuvant treatment decisions. Currently, clinical and histopathological factors such as stage, histotype, grade, depth of invasion, and lymphovascular space invasion are used to stratify patients into risk groups to guide surgical management and adjuvant therapy. However, these clinicopathological variables do not sufficiently predict patient outcomes. This highlights the need for improved classification of endometrial cancers to better predict patient prognosis.

DNA sequencing efforts, including by The Cancer Genome Atlas (TCGA) in 2013 (Kandoth et al., 2013), have resulted in the identification of four prognostically-significant molecular subtypes: copy-number low (CN low), and copy-number high (CN high), microsatellite instability-high (MSI-H, hypermutated), and POLE-mutated (ultramutated). CN low tumors are typically low-grade endometrioid carcinomas, which are defined by low genomic copy number alterations, low tumour mutation burden (TMB); they harbor very rare mutations in TP53, and frequent mutations in PTEN, PIK3CA, ARID1A, CTNNB1, and KRAS. In contrast, CN high tumors consist of serous carcinomas and some high-grade endometrioid carcinomas; these tumors have extensive copy number alterations, frequent mutations in TP53 and PIK3CA, and rare PTEN alerations. MSI-high tumors, which consist of low-grade and high-grade endometrioid carcinomas, have mismatch repair (MMR) protein defects superimposed on the mutations seen in CN low tumors. And finally, POLE-mutant tumors are defined by hotspot mutations in the exonuclease domain of DNA polymerase epsilon (POLE). POLE-mutant tumors have dramatically better progression-free survival, regardless of histotype and grade, whereas copy-number high tumors have the poorest outcome. The largest group (copy number low and MSI-high tumors) have intermediate prognosis. These prognostic molecular subtypes were identified by sequencing endometrioid and serous carcinomas, and it appears these molecular subtypes also exist in endometrial clear cell carcinoma (DeLair et al., 2017).

LAB BEST PRACTICE

The above TCGA classification is based on whole genome sequencing, which is not practical to employ in a clinical setting due to cost and time considerations. Therefore, a simple, inexpensive and clinically useable molecular classification system has been independently developed by two groups in Vancouver and the Netherlands. Termed ProMisE (Proactive Molecular Risk Classifier for Endometrial Cancer) by the Vancouver group, this molecular classification system uses three immunohistochemical (IHC) stains – p53, MMR proteins PMS2 and MSH6, and POLE exonuclease domain hotspot sequencing – as surrogate markers of the TCGA molecular subtypes (Figure 1) (Stelloo et al., 2016; Talhouk et al., 2015). The resultant molecular subtypes are termed termed p53 wild type / nonspecific molecular profile (p53wt/NSMP, surrogate of TCGA CN low), p53 abnormal (p53abn for staining patterns consistent with missense or null mutations, surrogate of CN high), MMR defective (MMR-D, surrogate of MSI-H), and POLE exonuclease domain mutant (POLE EDM, surrogate of POLE-mutated group). Kaplan?Meier survival analyses demonstrate that the simplified ProMisE molecular subgroups survival features mirror the TCGA genomic classification system (Figure 2). The classifier is effective regardless of histotype, which circumvents known problems of interobserver variation in the diagnosis and grading of endometrial cancers, particularly in high-grade tumors (Gilks et al., 2013; Han et al., 2013).

Figure 1. A simple, genomics-based clinical classifier for endometrial cancer (Talhouk et al., 2017)

This simplified molecular classification system has several additional advantages over the TCGA classification. It works on formalin-fixed paraffin-embedded tissue. p53 and MMR IHC stains are universally available, so implementing the system in a clinical setting only requires POLE hotspot sequencing. ProMisE can be used on biopsy samples to help guide surgical management (Talhouk et al., 2016). And most obviously, it is significantly cheaper and faster to use than the TCGA classifier.

Figure 2. The simplified ProMisE molecular subgroups survival mirror the TCGA genomic classification system (Talhouk et al., 2017)

Recently, it has been demonstrated that the two molecular subtypes with intermediate prognosis (p53wt/NSMP and MMR-D), which comprise the majority of patients, can be further stratified by additional IHC for L1CAM (L1 cell-adhesion molecule) expression (Bosse et al., 2014; Karnezis et al., 2017; Kommoss et al., 2018; Stelloo et al., 2016) to subgroups with significant different survival features: the L1CAM-positive subgroup at higher risk for fatal outcome, when compared to the L1CAM negative subgroup (Figure 3).

Figure 3. Kaplan–Meier overall survival analysis of ProMisE subgroups before (upper panel) and after (lower panel) stratifying for L1CAM status within the p53 wt/NSMP subgroup (Kommoss et al., 2018)

In summary, ProMisE is a powerful, simple, and clinically useable molecular classifier that can be integrated into daily practice for the management of endometrial cancer patients. Ongoing research efforts are focused on how to refine ProMisE with additional immunohistochemical markers to identify additional prognostically relevant groups, and how to optimally integrate the molecular data with conventional clinicopathological variables to best treat patients.

REFERENCES

  1. Bosse, T., Nout, R.A., Stelloo, E., Dreef, E., Nijman, H.W., Jurgenliemk-Schulz, I.M., Jobsen, J.J., Creutzberg, C.L., and Smit, V.T. (2014). L1 cell adhesion molecule is a strong predictor for distant recurrence and overall survival in early stage endometrial cancer: pooled PORTEC trial results. Eur J Cancer 50, 2602-2610.
  2. DeLair, D.F., Burke, K.A., Selenica, P., Lim, R.S., Scott, S.N., Middha, S., Mohanty, A.S., Cheng, D.T., Berger, M.F., Soslow, R.A., et al. (2017). The genetic landscape of endometrial clear cell carcinomas. J Pathol 243, 230-241.
  3. Gilks, C.B., Oliva, E., and Soslow, R.A. (2013). Poor interobserver reproducibility in the diagnosis of high-grade endometrial carcinoma. Am J Surg Pathol 37, 874-881.
  4. Han, G., Sidhu, D., Duggan, M.A., Arseneau, J., Cesari, M., Clement, P.B., Ewanowich, C.A., Kalloger, S.E., and Kobel, M. (2013). Reproducibility of histological cell type in high-grade endometrial carcinoma. Mod Pathol 26, 1594-1604.
  5. Kandoth, C., Schultz, N., Cherniack, A.D., Akbani, R., Liu, Y., Shen, H., Robertson, A.G., Pashtan, I., Shen, R., Benz, C.C., et al. (2013). Integrated genomic characterization of endometrial carcinoma. Nature 497, 67-73.
  6. Karnezis, A.N., Leung, S., Magrill, J., McConechy, M.K., Yang, W., Chow, C., Kobel, M., Lee, C.H., Huntsman, D.G., Talhouk, A., et al. (2017). Evaluation of endometrial carcinoma prognostic immunohistochemistry markers in the context of molecular classification. J Pathol Clin Res 3, 279-293.
  7. Kommoss, F.K., Karnezis, A.N., Kommoss, F., Talhouk, A., Taran, F.A., Staebler, A., Gilks, C.B., Huntsman, D.G., Kramer, B., Brucker, S.Y., et al. (2018). L1CAM further stratifies endometrial carcinoma patients with no specific molecular risk profile. Br J Cancer 119, 480-486.
  8. Siegel, R.L., Miller, K.D., and Jemal, A. (2019). Cancer statistics, 2019. CA Cancer J Clin 69, 7-34.
  9. Stelloo, E., Nout, R.A., Osse, E.M., Jurgenliemk-Schulz, I.J., Jobsen, J.J., Lutgens, L.C., van der Steen-Banasik, E.M., Nijman, H.W., Putter, H., Bosse, T., et al. (2016). Improved Risk Assessment by Integrating Molecular and Clinicopathological Factors in Early-stage Endometrial Cancer-Combined Analysis of the PORTEC Cohorts. Clin Cancer Res 22, 4215-4224.
  10. Talhouk, A., Hoang, L.N., McConechy, M.K., Nakonechny, Q., Leo, J., Cheng, A., Leung, S., Yang, W., Lum, A., Kobel, M., et al. (2016). Molecular classification of endometrial carcinoma on diagnostic specimens is highly concordant with final hysterectomy: Earlier prognostic information to guide treatment. Gynecol Oncol 143, 46-53.
  11. Talhouk, A., McConechy, M.K., Leung, S., Li-Chang, H.H., Kwon, J.S., Melnyk, N., Yang, W., Senz, J., Boyd, N., Karnezis, A.N., et al. (2015). A clinically applicable molecular-based classification for endometrial cancers. Br J Cancer 113, 299-310.
  12. Talhouk, A., McConechy, M.K., Leung, S., Yang, W., Lum, A., Senz, J., Boyd, N., Pike, J., Anglesio, M., Kwon, J.S., et al. (2017). Confirmation of ProMisE: A simple, genomics-based clinical classifier for endometrial cancer. Cancer 123, 802-813.