Mismatch repair markers in preoperative and operative endometrial cancer samples; expression concordance and prognostic value

Expression patterns of mismatch repair proteins in cervical cancer uncover independent prognostic value of MSH-2

OBJECTIVE

Although early-detected cervical cancer is associated with good survival, the prognosis for late-stage disease is poor and treatment options are sparse. Mismatch repair deficiency (MMR-D) has surfaced as a predictor of prognosis and response to immune checkpoint inhibitor(s) in several cancer types, but its value in cervical cancer remains unclear. This study aimed to define the prevalence of MMR-D in cervical cancer and assess the prognostic value of MMR protein expression.

METHODS

Expression of the MMR proteins MLH-1, PMS-2, MSH-2, and MSH-6 was investigated by immunohistochemical staining in a prospectively collected cervical cancer cohort (n=508) with corresponding clinicopathological and follow-up data. Sections were scored as either loss or intact expression to define MMR-D, and by a staining index, based on staining intensity and area, evaluating the prognostic potential. RNA and whole exome sequencing data were available for 72 and 75 of the patients and were used for gene set enrichment and mutational analyses, respectively.

RESULTS

Five (1%) tumors were MMR-deficient, three of which were of neuroendocrine histology. MMR status did not predict survival (HR 1.93, p=0.17). MSH-2 low (n=48) was associated with poor survival (HR 1.94, p=0.02), also when adjusting for tumor stage, tumor type, and patient age (HR 2.06, p=0.013). MSH-2 low tumors had higher tumor mutational burden (p=0.003) and higher frequency of (frameshift) mutations in the double-strand break repair gene RAD50 (p<0.01).

CONCLUSION

MMR-D is rare in cervical cancer, yet low MSH-2 expression is an independent predictor of poor survival.

Potential of molecular classification to guide fertility-sparing management among young patients with endometrial cancer

The traditional histological classification system for endometrial carcinoma falls short in addressing the disease's molecular heterogeneity, prompting the need for alternative stratification methods. Proactive Molecular Risk Classifier for Endometrial Cancer (ProMisE) has emerged as a clinically efficient tool to categorize endometrial cancers according to mismatch repair deficiency, POLE exonuclease domain mutations, and p53 expression. However, the application of this classification to fertility-sparing treatments remains unexplored, and current guidelines lack specificity in how it should be used. In this review, we summarize the available literature and establish the framework for future investigations focused on molecular profiling-based risk assessment of endometrial cancer, with the goal of utilizing precision medicine to optimally counsel patients seeking fertility-sparing treatment. While the available evidence is limited and of low quality, it does provide insights and frames future perspectives for managing fertility-sparing approaches on the basis of molecular subtypes. Evidence suggests that mismatch repair-deficient tumors are likely to recur despite progestin therapy, emphasizing the need for alternative treatments, with targeted therapies being a new landscape that still needs to be explored. Tumors with POLE mutations exhibit a favorable prognosis, but the safety of hysteroscopic resection alone requires further investigation. p53 abnormal tumors have an unfavorable prognosis, raising questions about their suitability for fertility-sparing treatment. Lastly, the no specific molecular profile (or p53 wild-type) tumors, while having a relatively good prognosis, are heterogeneous and require more precise biomarkers to effectively guide therapy for those with poorer prognoses. Addressing these research gaps will lead to more precise guidelines to ensure optimal selection for fertility-sparing treatment.

Prediction of Mismatch Repair Status in Endometrial Cancer from Histological Slide Images Using Various Deep Learning-Based Algorithms

The application of deep learning algorithms to predict the molecular profiles of various cancers from digital images of hematoxylin and eosin (H&E)-stained slides has been reported in recent years, mainly for gastric and colon cancers. In this study, we investigated the potential use of H&E-stained endometrial cancer slide images to predict the associated mismatch repair (MMR) status. H&E-stained slide images were collected from 127 cases of the primary lesion of endometrial cancer. After digitization using a Nanozoomer virtual slide scanner (Hamamatsu Photonics), we segmented the scanned images into 5397 tiles of 512 × 512 pixels. The MMR proteins (PMS2, MSH6) were immunohistochemically stained, classified into MMR proficient/deficient, and annotated for each case and tile. We trained several neural networks, including convolutional and attention-based networks, using tiles annotated with the MMR status. Among the tested networks, ResNet50 exhibited the highest area under the receiver operating characteristic curve (AUROC) of 0.91 for predicting the MMR status. The constructed prediction algorithm may be applicable to other molecular profiles and useful for pre-screening before implementing other, more costly genetic profiling tests.

Mismatch repair gene MSH6 correlates with the prognosis, immune status and immune checkpoint inhibitors response of endometrial cancer

Objective

Many patients treated with immune checkpoint inhibitors (ICIs) developed primary or secondary drug resistance for unknown reasons. This study investigates whether mismatch repair (MMR) genes are responsible for this therapeutic restriction.

Methods

We obtained the transcriptional, clinical and single nucleotide polymorphism data for endometrial cancer (EC) from The Cancer Genome Atlas and the immunophenoscore data of EC from The Cancer Immunome Atlas, then analyzed in R to evaluate the relationship between MMR genes and clinicopathological features, prognosis, immune infiltration, immune checkpoint expression and responsiveness to ICIs in EC. We used differentially expressed genes in the MSH6 high and low expression groups to conduct GO and KEGG analyses to explore the impact of MSH6 on the biological functions of EC. Finally, we verified the bioinformatics results with in vitro experiments.

Results

Our analyses showed that compared with the high MSH6 expression group, the low MSH6 expression group had better survival outcomes and less aggressive clinicopathological features. In the multivariate Cox analysis, MSH6 was the only independent risk factor that could predict the prognosis of EC. Besides, the low MSH6 expression group also had a higher immune score, more active immune infiltration and higher immune checkpoint expression, resulting in better responsiveness to ICIs treatment, consistent with the enrichment of GO terms and KEGG pathways related to immune response in this group. Meanwhile, the GO and KEGG enrichment results of the MSH6 high expression group were associated with cell cycle, DNA damage repair and tumorigenesis. To exclude the influence of MSH6 mutations, we performed the previous analyses on the MSH6 wild-type tumor samples and obtained consistent results. In vitro experiments also confirmed that after knocking down MSH6 in endometrial cancer cells, their proliferation, migration and invasion abilities were weakened, while the expression levels of PD-L1 and PD-L2 were elevated. In comparison, overexpression of MSH6 showed an opposite trend.

Conclusion

Reduced MSH6 expression could serve as a potential biomarker for predicting better prognosis, active immune status, higher immune checkpoint expression level and better responsiveness to ICIs treatment in EC. MSH6 may become a potential target for treating solid tumors.

A retrospective study of consistency between immunohistochemistry and polymerase chain reaction of microsatellite instability in endometrial cancer

Objectives

Identification of endometrial cancers (EC) with mismatch repair deficiency (dMMR) or microsatellite instability-high (MSI-H) is essential for Lynch syndrome screening and treatment stratification. We aimed to assess the utility of immunohistochemistry (IHC) staining for MMR protein expression and polymerase chain reaction (PCR)-based MSI assays in EC and the correlation between MMR/MSI status and various clinicopathological parameters.

Methods

We reviewed the clinical and pathological information of 333 patients with EC. MMR protein expression was assessed as retained or lost to determine MMR status by IHC staining, and MSI status was identified by PCR capillary electrophoresis (PCR-CE) testing with a National Cancer Institute (NCI) panel. The correlation of MMR/MSI status with clinicopathological features was determined by statistical analysis. Discrepant results were further analyzed using an alternative PCR-CE MSI (Promega panel) method, MLH1 promoter methylation assays, and next-generation sequencing (NGS).

Results

Among the EC patients, the overall percentage of dMMR was 25.2%, and the overall percentage of MSI-H was 24%. Among the dMMR patients, 50 (59.5%) showed loss of MLH1 and PMS2 expression, 19 (22.6%) loss of MSH2 and MSH6 expression, and seven (8.3%) and eight (9.5%) loss of PMS2 and MSH6 expression, respectively. The dMMR subgroup was significantly younger than the pMMR subgroup, especially for <60-years-old patients (p = 0.038). In addition, we identified a strong correlation between MMR/MSI status and high-grade endometrioid or nonendometrioid components (p = 0.004 or p = 0.003). IHC staining and PCR-CE assay results showed a high level of overall concordance (98.8%, Cohen's κ = 0.98). Four patients were found to have dMRR/MSS in both examinations. We reanalyzed them with additional methods. One case showed MLH1 promotor methylation, and the other three cases harbored MSH6 germline pathogenic variations. One of the cases with MSH6 deficiency was reanalyzed as MSI-H by alternative PCR-CE assay or NGS testing.

Conclusions

This study indicates that the combined use of MMR-IHC and PCR-CE MSI analyses may effectively avoid misdiagnoses of EC patients with dMMR/MSI-H. However, use of PCR-CE alone to evaluate MMR/MSI status may lead to missed diagnosis, especially for EC patients with MSH6 deficiency and presenting MSS.

Epigenetic MMR defect identifies a risk group not accounted for through traditional risk stratification algorithms in endometrial cancer

Purpose

We sought to evaluate the contribution of mismatch repair (MMR) status to traditional risk stratification algorithms used to predict nodal involvement and recurrence in a large single-institution cohort.

Methods

Endometrioid endometrial cancer (EC) cases from 2014-2020 were evaluated. MMR immunohistochemistry (IHC) was performed universally. Uterine factors assessed in the Mayo criteria were used to retrospectively classify patients as low or high risk for lymphatic spread. Patients were classified according to risk for recurrence using GOG 99 and PORTEC criteria. Associations were evaluated using chi-square and t-tests and contributing factors assessed using logistic regression models.

Results

1,514 endometrioid EC were evaluated; 392 (25.9%) were MMR (MMR) deficient of which 80.4% of MMR defects were associated with epigenetic silencing of MLH1. Epigenetic MMR defects were significantly more likely to be high risk for lymph node (LN) metastasis based on Mayo criteria (74.9% vs 60.6%, p=<0.001) and with the presence of LN metastasis (20.3 vs 10.5%, p=0.003) compared to MMR proficient tumors. Tumors with epigenetic MMR defects were significantly more likely to be classified as high or high intermediate risk using GOG99 and PORTEC criteria. Furthermore, cases with epigenetic MMR defects classified as low or low intermediate risk were significantly more likely to recur (GOG99 p=0.013; PORTEC p=0.008) and independently associated with worse disease-free survival (DFS). MMR status was found to be independently associated with worse DFS (HR 1.90; 95% CI 1.34-2.70; p=0.003) but not overall survival.

Conclusion

While MMR deficient EC has been associated with poor prognostic features in prior reports; we demonstrate that only epigenetic MMR defects have poorer outcomes. Epigenetic MMR defect were independently associated with lymph node metastasis after controlling for risk criteria. Epigenetic MMR deficiency was found to be an independent predictor of recurrence beyond the factors considered in traditional risk stratification algorithms. Traditional uterine-based risk stratification algorithms may not fully reflect the risk for recurrence in MMR deficient tumors. Consideration should be given to implementing MMR status and MLH1 hypermethylation alongside traditional risk stratification algorithms. Performing MMR IHC on preoperative pathologic specimens may aid in risk stratification and patient counseling.