LAUSR

The purpose of cancer staging is to classify cancers into prognostic groups and to allow for comparison of treatment results and survival between patients and institutions. Staging for gynecologic cancers is based on extent of disease and metastasis; it was historically determined by physical examination and is now based on surgical and histologic examination of tumor specimens. Although extent of disease is considered the most important predictor of recurrence and survival, current staging usually includes histologic type but not molecular markers that are associated with tumor aggressiveness, response to therapy, and prognosis. Molecular technology offers the opportunity to characterize and predict the behavior of cancers even better. Endometrial cancer is currently is staged on the basis of anatomic location and extent of the disease. Unfortunately, the stage of disease is not perfect in predicting recurrence or cancer-related survival, because some patients with early stage cancer recur early and die, whereas some with advanced-stage disease may have long-term survival despite extensive disease. Although the histologic appearance of the tumor is also an important prognostic variable, interobserver reproducibility in the pathologic diagnosis of high-grade subtypes (ie, endometrioid, serous, clear cell, and carcinosarcoma) is limited. The advent of molecular profiling affords an opportunity to enhance current prognostic models by combining anatomic, histologic, and molecular features of the index cancer. In the current study by Talhouk et al, those investigators confirmed a prognostic panel based on multiplatform molecular subtype classifications identified by The Cancer Genome Atlas (TCGA) (ie, polymerase e [POLE], microsatellite instability [MSI], copy-number [CN]-low, and CN-high). Although the Proactive Molecular Risk Classification Tool for Endometrial Cancers (ProMisE) panel is not identical in reflecting the various molecular features of TCGA subtypes, it does provide a practical representation of prognosis using commonly used assays. In TCGA, the assessment of MSI was performed by comparing allelic profiles of matched tumor and germline DNA using 7 markers (BAT-25, BAT-26, NR-21, NR-24, MONO-27, Penta C, and Penta D), whereas the ProMisE assay used mismatch-repair (MMR) immunohistochemistry testing for mutL homolog 1 (MLH1), mutS homolog 2 (MSH2), MSH6, and PMS1 homolog 2 (PMS2). Instead of Affymetrix single nucleotide polymorphism (SNP) arrays (Affymetrix, Santa Clara, Calif), which TCGA used to assess CN variation, the ProMisE investigators used immunohistochemistry to assess tumor protein 53 (p53) mutation as a surrogate marker for CN-high patients. Sequencing of POLE was the third element of the ProMisE panel. In the testing, tumors are first assessed for MMR, then POLE mutations, and finally abnormal p53 expression. Some cases may have more than 1 molecular feature (approximately 3%) but get categorized on the first feature in the diagnostic tree. The authors previously reported correlations of the discovery panel with outcomes and have reported their current findings using an independent confirmation cohort as well as the full cohort (discovery 1 confirmation cohorts) in the supporting materials. Patients with abnormal p53 expression appear to be the principal determinant of poor cancer-related outcome. Identification of a prognostic panel for endometrial cancer ideally would involve a discovery analysis aimed at the identification of any known and unknown biomarkers associated with outcome and then discerning what combinations of molecular alterations are most accurately associated with outcome. Molecular alterations characterizing the ProMisE panel were not developed through this type of unbiased exploratory analysis but, instead,