Predictive accuracy of hormone receptors in conservatively treated endometrial hyperplasia and early endometrioid carcinoma

Fertility-Sparing Approaches in Atypical Endometrial Hyperplasia and Endometrial Cancer Patients: Current Evidence and Future Directions

Endometrial cancer (EC) is the fourth most common cancer in women in developed countries. Although it is usually diagnosed in postmenopausal women, its incidence has increased in young women, as well in recent decades, with an estimated rate of 4% in those under 40 years of age. Factors involved in this increase, particularly in resource-rich countries, include delayed childbearing and the rise in obesity. The new molecular classification of EC should help to personalize treatment, through appropriate candidate selection. With the currently available evidence, the use of oral progestin either alone or in combination with other drugs such as metformin, levonorgestrel-releasing intrauterine devices and hysteroscopic resection, seems to be feasible and safe in women with early-stage EC limited to the endometrium. However, there is a lack of high-quality evidence of the efficacy and safety of conservative management in EC. Randomized clinical trials in younger women and obese patients are currently underway.

Dusp6 immunohistochemistry is associated with the response of atypical endometrial hyperplasia and early endometrial cancer to conservative treatment

OBJECTIVE

Dual-specificity phosphatase 6 (Dusp6) was proposed as a predictive marker of response of atypical endometrial hyperplasia (AEH) and early endometrial cancer (EEC) to conservative treatment. However, its predictive accuracy has never been calculated. We aimed to define it in conservatively treated AEH and EEC.

METHODS

All patients <45 years with AEH or EEC and conservatively treated with hysteroscopic resection + LNG-IUD insertion from 2007 to 2018 were retrospectively assessed. Dusp6 immunohistochemical expression was assessed and dichotomized as "strong" vs "weak". Relative risk (RR) for "no regression" and "recurrence" or AEH/EEC was calculated. Predictive accuracy was calculated as sensitivity, specificity, positive and negative predictive values (PPV, NPV) and area under the curve (AUC) on receiver operating characteristic curve.

RESULTS

Thirty-six women were included. Weak Dusp6 immunohistochemical expression was significantly associated with increased risk of resistance to treatment, with a RR = 16 (P = 0.0074); predictive accuracy analysis showed sensitivity = 80%, specificity = 90%, PPV = 57.1%, NPV = 96.4%, AUC = 0.85. A weak Dusp6 expression was not significantly associated with the risk of recurrence after an initial regression (RR = 0.4; P = 0.53).

CONCLUSION

Weak Dusp6 expression appears as a significant predictor of resistance of AEH/EEC to fertility-sparing treatment, with moderate predictive accuracy. Weak Dusp6 expression is significantly associated with resistance of atypical endometrial hyperplasia or early endometrial cancer to fertility-sparing treatment, with moderate predictive accuracy.

Fertility-Sparing Treatment for Atypical Endometrial Hyperplasia and Endometrial Cancer: A Cochrane Systematic Review Protocol

INTRODUCTION

Endometrial cancer is the fifth most common cancer in women and atypical endometrial hyperplasia is a precancerous lesion. Obesity is an important risk factor for endometrioid endometrial adenocarcinoma and endometrial hyperplasia. Progesterone is recommended as first-line treatment in endometrial cancer or atypical endometrial hyperplasia in women who wish to preserve fertility, but optimal treatment schedules have not been defined. Metformin or bariatric surgery may also be useful in these women. The effectiveness and safety of fertility-preserving treatments being used for women with atypical endometrial hyperplasia and stage IA grade 1 endometrial cancer is unclear. Therefore, the systematic review aims to determine this point.

METHODS

We will search the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, trial registers, conference proceedings, abstracts, cooperative trial groups and reference lists. We will include randomised controlled trials (RCTs) that compare fertility-preserving therapy including orally administered progesterone versus a levonorgestrel-releasing intrauterine system (IUS), metformin, other pharmacological interventions or bariatric surgery, and any of these interventions with womb-removing surgery. Quasi-randomised trials, non-randomised trials and cohort studies will be included. Two review authors will independently assess study eligibility and risk of bias and extract data. The primary outcomes are complete pathologic response and live birth rate. Secondary outcomes include overall survival, progression-free survival, pregnancy rate, need for hysterectomy, adverse events, psychological symptoms and quality of life.

PLANNED OUTCOMES

This review aims to clarify the effectiveness and risks of fertility-preserving treatments, including complete pathologic response rate, live birth rates, need for surgical treatment, adverse events, psychological symptoms and quality of life. The broad scope of the review includes the use of progesterone, metformin to reverse insulin resistance, and bariatric surgery or operative hysteroscopy.

RESULTS

The results may help to determine the optimal fertility-sparing treatment in endometrial cancer and atypical endometrial hyperplasia.

SYSTEMATIC REVIEW REGISTRATION

Prospero 2019 number CRD42019145991.

Predictive Accuracy of Progesterone Receptor B in Young Women with Atypical Endometrial Hyperplasia and Early Endometrial Cancer Treated with Hysteroscopic Resection plus LNG-IUD Insertion

STUDY OBJECTIVE

The immunohistochemical expression of isoform B of the progesterone receptor (PRB) has shown promising results in predicting the response of atypical endometrial hyperplasia (AEH) and early endometrial cancer (EEC) to conservative treatment. We aimed to calculate the accuracy of PRB as a predictive marker of conservative treatment outcome in AEH or EEC.

DESIGN

Retrospective cohort study.

SETTING

University of Naples Federico II, Naples, Italy.

PATIENTS

Thirty-six consecutive premenopausal women <45 years of age with AEH (n = 29) or EEC (n = 7) conservatively treated from January 2007 to June 2018 were retrospectively assessed.

INTERVENTIONS

All patients had been treated with hysteroscopic resection plus levonorgestrel-releasing intrauterine device insertion and followed for at least 1 year. The immunohistochemical expression of PRB was separately assessed in the glands and stroma of the lesion and dichotomized as "weak" or "normal."

MEASUREMENT AND MAIN RESULTS

The treatment outcomes considered were (1) treatment failure (i.e., a combined outcome including no regression or recurrence); (2) no regression; and (3) recurrence. The predictive accuracy of PRB immunohistochemistry was assessed by calculating sensitivity (SE), specificity (SP), and area under the receiver operating characteristic curve (AUC). A weak glandular PRB expression showed SE = 70%, SP = 77%, and AUC = 0.74 for treatment failure; SE = 66.7%, SP = 70%, and AUC = 0.68 for no regression; and SE = 75%, SP = 68.8%, and AUC = 0.72 for recurrence. A weak stromal PRB expression showed SE = 100%, SP = 53.8%, and AUC = 0.77 for treatment failure; SE = 100%, SP = 46.7%, and AUC = 0.73 for no regression; and SE = 100%, SP = 43.8%, and AUC = 0.72 for recurrence.

CONCLUSION

A weak stromal PRB expression is a highly sensitive predictive marker of both no response and recurrence of AEH and EEC conservatively treated.

Diagnostic accuracy of p53 immunohistochemistry as surrogate of TP53 sequencing in endometrial cancer

Aberrant p53 immunohistochemical expression is used to identify the copy-number-high/TP53-mutant subgroup of endometrial cancer (EC). We aimed to determine the diagnostic accuracy of p53 immunohistochemistry as surrogate for TP53 sequencing through a systematic review and meta-analysis. Electronic databases were searched from their inception to June 2019. All studies assessing p53 expression and TP53 mutations in EC were included. Diagnostic accuracy was assessed based on area under the curve (AUC). Immunohistochemical criteria used to define aberrant p53 expression were "overexpression" and "overexpression or complete absence". Subgroup analysis was based on the sequencing technique adopted (Polymerase Chain Reaction + sequencing, or next generation sequencing, NGS). Thirteen observational studies with 727 endometrial cancers were included. Both "overexpression" and "overexpression or complete absence" showed high diagnostic accuracy (AUC = 0.9088 and 0.9030, respectively). The subgroup with "overexpression" and NGS showed the best results, with very high diagnostic accuracy (AUC = 0.9927). In conclusion, immunohistochemistry for p53 is a highly accurate surrogate of TP53 sequencing. Overexpression of p53 in ≥70-80% showed the best accuracy in predicting TP53 mutations. Further studies in this field should adopt optimized immunohistochemical procedures and take into account less common p53 patterns (e.g. cytoplasmic expression).

Impact of endometrial carcinoma histotype on the prognostic value of the TCGA molecular subgroups

BACKGROUND

The Cancer Genome Atlas (TCGA) identified four prognostic subgroups of endometrial carcinoma: copy-number-low/p53-wild-type (p53wt), POLE-mutated/ultramutated (POLEmt), microsatellite-instability/hypermutated (MSI), and copy-number-high/p53-mutated (p53mt). However, it is still unclear if they may be integrated with the current histopathological prognostic factors, such as histotype.

OBJECTIVE

To assess the impact of histotype on the prognostic value of the TCGA molecular subgroups of endometrial carcinoma.

METHODS

A systematic review and meta-analysis was performed by searching 7 electronic databases from their inception to April 2019 for studies assessing prognosis in all TCGA subgroups of endometrial carcinoma. Pooled hazard ratio (HR) for overall survival (OS) was calculated in two different groups ("all-histotypes" and "endometrioid"), using p53wt subgroup as reference standard; HR for non-endometrioid histotypes was calculated indirectly. Disease-specific survival and progression-free survival were assessed as additional analyses.

RESULTS

Six studies with 2818 patients were included. In the p53mt subgroup, pooled HRs for OS were 4.322 (all-histotypes), 2.505 (endometrioid), and 4.937 (non-endometrioid). In the MSI subgroup, pooled HRs were 1.965 (all-histotypes), 1.287 (endometrioid), and 6.361 (non-endometrioid). In the POLEmt subgroup, pooled HRs were 0.763 (all-histotypes), 0.481 (endometrioid), and 2.634 (non-endometrioid). Results of additional analyses were consistent for all subgroups except for non-endometrioid POLEmt carcinomas.

CONCLUSION

Histotype of endometrial carcinoma shows a crucial prognostic value independently of the TCGA molecular subgroup, with non-endometrioid carcinomas having a worse prognosis in each TCGA subgroup. Histotype should be integrated with molecular characterization for the risk stratification of patients in the future.