Abstractabstract
Endometrial hyperplasia is of great clinical importance as a precursor to endometrial adenocarcinoma. It is also clinically important to clearly distinguish between endometrial hyperplasia, precancerous lesions and malignant tumors. Confusion may lead to under- or over-treatment because we need to treat the disease accordingly. Each type of precancerous lesion has a different clinical management, so we need a pathological description that reflects the diagnostic criteria and clearly differentiates the clinicopathological types of the different lesions. The “endometrial intraepithelial neoplasia” diagnosis and treatment system was born out of this desire. The system incorporates the advantages of the previous pathological diagnosis, while some modifications have been made. The new system is still based on the 4-category pathological diagnosis model proposed by the World Health Organization in 1994 for non-malignant endometrial diseases (in this model, atypical hyperplasia is equated with precancerous lesions). It is unclear whether diagnostic curettage or endometrial aspiration is superior in the diagnosis of precancerous lesions and whether they are combined with carcinogenesis; however, direct hysteroscopic sampling is clearly the most sensitive means of extraction. We recommend that surgery should be the first choice for patients with endometrial intraepithelial neoplasia when clinical circumstances allow. This is because total hysterectomy not only provides a definitive assessment of the disease (whether it is combined with cancer), but also provides effective treatment of precancerous lesions. However, when patients are intolerant to surgery or require preservation of reproductive function, systemic or local progestin application is a commonly used hand alternative surrogate therapy, but its certainty needs further confirmation.
Conclusions and recommendations
Sensitive and precise diagnosis of endometrial precancerous lesions may reduce the likelihood of their development into invasive cancer. Based on available data and expert opinion, the American College of Obstetricians and Gynecologic Oncologists and the Society of Gynecologic Oncology have developed the following consensus.
● The current pathologic description system for endometrial intraepithelial neoplasia appears to be superior to the WHO 94 version of the pathologic description system. Each type of precancerous lesion has a different clinical management, so we need a terminology that reflects the diagnostic criteria and clearly differentiates the different clinicopathological types of lesions. For this purpose, the “endometrial intraepithelial neoplasia” diagnostic system was developed, which combines and modifies the previous pathological diagnostic criteria. The new pathological criteria are based on the 1994 World Health Organization model of four pathological types of non-malignant endometrial diseases (in which atypical hyperplasia is equated with precancerous lesions). The term “endometrial intraepithelial neoplasia” is a better professional descriptor (better than “endometrial atypical hyperplasia”).
For histologic sampling, we recommend direct hysteroscopy (although not mandatory) to obtain as much (small and scattered) lesion tissue as possible and to reduce background (normal endometrial tissue) interference. This will give us a better chance to confirm the diagnosis of a true precancerous lesion and to clarify whether it is combined with endometrial cancer. When clinical circumstances allow, total hysterectomy provides a definitive assessment (for combined cancer) of patients with endometrial intraepithelial neoplasia; and effective treatment of precancerous lesions.
● Subtotal hysterectomy, uterine fractionation, and endometrial resection are not indicated for patients with endometrial intraepithelial neoplasia.
● Systemic or local progestin is a common alternative therapy to hysterectomy, but its validity needs to be further confirmed; it is generally used only in patients who are intolerant to surgery or require preservation of fertility.
In patients with endometrial intraepithelial neoplasia not treated surgically and who opt for hormonal therapy, subsequent surveillance follow-up should include a series of endometrial biopsies every 3-6 months. However, the exact frequency of follow-up examinations has not been determined.
Background
Endometrial hyperplasia is clinically important because it often progresses to adenocarcinoma.Type I endometrioid adenocarcinoma is characterized by a preexisting lesion: endometrial intraepithelial neoplasia. Proliferation of the endometrial glandular epithelium occurs only in response to estrogen stimulation without progesterone antagonism. This is a physiologically distinct process from true precancerous lesions and cancers and is fundamentally due to sustained long-term hormonal exposure. It is clinically relevant to recognize this, as otherwise under- or over-treatment can occur during clinical treatment. The core mission of this society is the classification and treatment of endometrial hyperplasia. Gynecologists should be aware of the differences between the two aforementioned pathologic diagnostic systems and know that the endometrial intraepithelial neoplasia pathologic description system is superior to the 94th edition of the WHO pathologic description system. Each type of precancerous lesion has a different clinical management, so we need a terminology that reflects the diagnostic criteria and clearly distinguishes the different clinicopathological types of lesions. For this purpose, the “endometrial intraepithelial neoplasia” diagnostic system was developed, which combines and modifies the previous pathological diagnostic criteria. The new pathological criteria were established on the basis of the 1994 World Health Organization’s model of four pathological types of non-malignant endometrial diseases (in which atypical hyperplasia is equated with precancerous lesions). “Endometrial intraepithelial neoplasia” (rather than “endometrial atypical hyperplasia”) is the preferred pathology term and is used in this article.
Classification system of endometrial hyperplasia
There are two current pathological description systems for the terminology of endometrial precancerous lesions: 1) the 94th edition of the WHO system and 2) the system of endometrial intraepithelial neoplasia initiated by the International Endometrial Lesion Collaboration.2 The 94th edition of the WHO system gives four histopathological types, based on the complexity and heterogeneity of the proliferating glands. This system classifies precancerous lesions into four categories according to the risk of cancer – simple hyperplasia, complex hyperplasia, simple hyperplasia combined with nuclear heterogeneity, and complex hyperplasia combined with nuclear heterogeneity. This classification system is actually descriptive of the morphology of the lesion and has a heavy subjective component of the diagnostician’s thinking; therefore, it has poor reproducibility in terms of consistent pathological findings for each case. In addition, this system does not recommend a corresponding management plan for each pathological subtype. This old descriptive system should be used more for the description of pathological patterns only, and we should apply another descriptive system for clinical management.
The pathological description of endometrial intraepithelial neoplasia refers to precancerous endometrial lesions as “endometrial intraepithelial neoplasia”. In this system, endometrial lesions are pathologically classified into three categories: 1) benign – benign endometrial hyperplasia; 2) precancerous – endometrial intraepithelial neoplasia; 3) malignant – endometrial adenocarcinoma (endometrial cancer). -endometrial adenocarcinoma (endometrium-like morphology with good differentiation) (see Tables 1 and 2). By applying this system of classification to endometrial biopsy tissue, the pathologist can provide the clinician with a specific disease classification pattern that informs the clinician’s treatment decisions. The findings of several retrospective studies and one prospective study confirm that the application of this pathology description system is prognostically indicative of disease. Two of these studies also suggest that the reproducibility of pathologic findings across pathologists is higher with this pathologic description system than with the 94-year-old WHO diagnostic pathology system. This is the main reason for the preference of gynecologists for this system.
Diagnosis of precancerous lesions: consideration of endometrial biopsy sampling
Sensitive and specific diagnosis of endometrial lesions, along with clarification of whether they are combined with cancer, is a prerequisite for the development of a treatment plan for patients with precancerous lesions. It is not possible to determine whether the endometrial lesion is cancerous by endometrial aspiration alone. Approximately 40% of patients with endometrial intraepithelial neoplasia concluded by Dilation and Curettage (D&C) are diagnosed with endometrial cancer after hysterectomy.
It is unclear whether D&C is superior to endometrial aspiration in diagnosing precancerous endometrial lesions and in determining the presence of carcinogenesis. Both have limitations in terms of tissue sampling. Approximately 60% of D&C procedures yield less than half of the total endometrial tissue volume. If hysterectomy is chosen as the management of the disease, the mode of retrieval is irrelevant because hysterectomy alone excludes the possibility of missing endometrial cancer due to retrieval. Diagnostic curettage and endometrial aspiration have the same cancer detection rate in patients with abnormal vaginal bleeding. A single-center retrospective series found that diagnostic curettage in the diagnosis of endometrial intraepithelial neoplasia had a lower rate of missed cancer than endometrial aspiration (evidence from a large number of hysterectomies, where the rate was 27% versus 46%). When the diseased tissue mass is large and affects the shape of the endometrial cavity, endometrial aspiration instrumentation is compromised resulting in inadequate retrieval and therefore less valuable disease assessment. In this case, more sensitive sampling is performed under direct hysteroscopic view. Therefore, histological sampling is recommended (although not necessary) under direct hysteroscopic view to obtain as much (small or scattered) diseased tissue as possible, with less background (normal endometrial tissue) interference. This will give us a better chance to confirm the diagnosis of a true precancerous lesion and to clarify the presence of a combined cancerous lesion. It also affects the accurate assessment of cancer risk if the amount of tissue obtained is low. Therefore, just as we evaluate cervical tissue sampling, we should also evaluate the accuracy of endometrial sampling, and this should be included in the newly established diagnostic evaluation system.
Diagnosis of endometrial cancer in women with postmenopausal bleeding
The application of transvaginal ultrasound scanning techniques to women with abnormal postmenopausal bleeding has a good negative predictive rate. When the patient’s endometrial thickness is found to be ≤4 mm, endometrial sampling is not necessary, and therefore the prevalence of endometrial malignancy is extremely low at that time. When the endometrial thickness is >4 mm, further investigations such as hysterosalpingography, formal hysteroscopy or endometrial biopsy should be performed. When the endometrial thickness exceeds 4 mm, ultrasonography alone is not adequate. In postmenopausal women with no abnormal symptoms, there is no conclusive evidence as to how significant the endometrial thickness is when it exceeds 4 mm; further testing is usually not recommended. The assessment of endometrial thickness by ultrasound scan for malignant disease is limited to postmenopausal women with abnormal bleeding.
Management of endometrial intraepithelial neoplasia
Our initial goal in patients with newly diagnosed endometrial intraepithelial neoplasia is to identify any combined adenocarcinoma and to design a rational treatment plan to deal with the insidious, mysterious, or possible future cancer and to try to prevent it from being missed or from being detected only when it has progressed to cancer. Hysterectomy is an effective treatment option for patients with endometrial intraepithelial neoplasia, but the results of non-surgical treatment are not well evaluated by the current data.
Surgical evaluation and treatment options
Total hysterectomy in patients with endometrial epithelial neoplasia offers definite advantages, both in the treatment of precancerous lesions and in the definitive evaluation of the disease, if the clinical situation permits. Currently, patients with endometrial intraepithelial neoplasia choose hysterectomy that can be performed transabdominally, transvaginally, or minimally invasive (or combined with adnexal resection).
Supracervical hysterectomy (subtotal hysterectomy), uterine fractionation and endometrial resection are not indicated for patients with endometriotic endothelial neoplasia. Subtotal hysterectomy is not recommended because it is not possible to fully measure the occurrence of potential cancerous lesions. Removal of the lower part of the cervix and body of the uterus allows for incidental detection of combined carcinomas and staging of the disease, which eliminates the risk of missing lesions. Patients with probable or proven uterine malignancy are contraindications to uterine fractionation. Even if the correct surgical procedure is used as recommended above, the patient should be informed that a secondary procedure (full staging procedure) should be performed once the cancerous lesion is detected.
The surgical approach and extent will vary according to the intraoperative exploration findings and pathological conclusions. Intraoperative evaluation includes incision of the specimen to look for evidence of tumor and to see the depth of tumor muscle infiltration. If invasive carcinoma is considered preoperatively, the pathologist should consider the representativeness of the frozen pathology results of the material taken. The surgeon should also consider the possibility of inconsistency between the frozen pathology and the paraffin pathology (although this is highly unlikely to occur).
Intraoperative rapid cryopathology results help determine the need for a full staging procedure. The concordance between rapid pathology and paraffin pathology in assessing the type of disease pathology, staging, and depth of muscle infiltration was 97.5%, 88%, and 98.2%, respectively. Rapid cryopathology was somewhat more accurate for diagnosis in high-risk disease. If a gynecologic oncologist is not present, the best strategy is to await the final paraffin pathology results and use them to select which patients should undergo full-stage surgery.
For the vast majority of patients, full staging + pelvic and abdominal aortic lymph node dissection for endometrial intraepithelial neoplasia leads to overtreatment and increases the risk of surgery. The percentage of preoperative biopsies with endometrial intraepithelial neoplasia and postoperative high-risk endometrial cancer (advanced stage, deep myeloid infiltration) is approximately 10%. Routine pelvic and para-aortic lymph node dissection in patients with a diagnosis of endometrial intraepithelial neoplasia alone also makes the surgical risk in these patients equivalent to that of a fully staged procedure. Total hysterectomy (with or without adnexal resection), with abdominal irrigation fluid and a gynecologic oncologist involved in the staging evaluation, is the most appropriate surgical management of endometrioid endothelial tumors.
In some cases, transvaginal removal of the uterus is technically challenging, such as the simultaneous removal of both ovaries. Also, full surgical staging is not practical for transvaginal surgery. Surgical simultaneous bilateral adnexal resection is also unnecessary, especially in premenopausal women. Bilateral adnexal resection in premenopausal or perimenopausal women without definite evidence of gynecologic malignancy increases the morbidity and mortality of women. The risk of residual potential cancer should be weighed against the risk of postoperative menopause, and secondary surgical removal of the ovaries should be chosen only if there is residual cancer.
Non-surgical treatment
Women with fertility requirements or those who cannot be treated surgically due to comorbidities should opt for non-surgical treatment. The goal of conservative treatment for women with fertility requirements is to completely remove the lesion, reverse the lesion to normal endometrium, and prevent the appearance of cancerous lesions. For patients who are inoperable due to comorbidities, the goals of treatment are to stabilize the lesion, remove as many risk factors as possible, and shift to a chronic disease treatment model. Currently, the conservative approach to treatment is limited to hormonal therapy.
Several studies have evaluated the effectiveness of hormonal therapy in the treatment of endometrial hyperplasia and have shown that progestin therapy is beneficial and that drug toxicity is controlled within permissible limits. Progestin therapy is a better option for women who require preservation of reproductive function, who have endometrial hyperplasia or precancerous lesions and for older patients who have endometrial intraepithelial neoplasia and/or early endometrial cancer but are inoperable due to severe medical comorbidities.
Progestins can counteract the mitogenic effects induced by estrogen and induce secondary differentiation. However, to date, there are no norms or guidelines for progestin use in published academic research journals. Only some studies have been seen that describe the clinical significance of progestins in endometrial hyperplasia.
Medroxyprogesterone acetate and megestrol acetate are the two most commonly used drugs in progestin therapy, and they are available in different dosage forms and regimens (see Table 3). Reversal of endometrial hyperplasia (including simple, complex and atypical hyperplasia) was seen in 80-90% of patients given medroxyprogesterone acetate (orally, 10 mg/d, 12-14 ds/m or transvaginal administration of micronized progesterone, 100 mg/d, 12-14 ds/m). However, endometrial intraepithelial neoplasia greatly increases the chance of failure of progestin therapy; also, the chance of disease progression to cancer is significantly higher than in hyperplastic lesions.
Systemic or local progestin therapy is the usual alternative to hysterectomy, the exactness of which has yet to be clearly proven; it is generally used in women who are intolerant of surgery or who require preservation of fertility. In addition to systemic hormonal agents, the feasibility of treatment with intrauterine devices (IUDs) that release progesterone has been investigated. IUDs that release levonorgestrel for 5 years are a good alternative to oral progesterone therapy. The effect of locally applied progesterone on the endometrium is much stronger (it is several times more relevant) than systemic administration. The number of cases selected in most trials investigating this content is small, but one study included 105 women. This study showed that levonorgestrel-containing IUDs in women with endometrial hyperplasia (including simple, complex, and atypical hyperplasia) had an effective cure rate of 90%, with 67% of cases being atypical hyperplasia. A systematic review and Meta-analysis found that oral progesterone treatment given to women with atypical hyperplasia (total of 189 female patients in 14 studies) had an overall effectiveness rate of 69% (CI:95%, [58,93]). And a combination of seven studies applying levonorgestrel IUDs showed that 90% of 36 female patients suffering from atypical hyperplasia achieved satisfactory results (CI:95%, [62,100]) [35].
Several issues remain to be resolved in the hormonal treatment of endometrial intraepithelial neoplasia, such as: effective therapeutic dose and duration of application, whether the drug should be applied cyclically or continuously. The rules for follow-up after hormone therapy are also not clear. We also lack appropriate criteria for assessing clinical and histological changes after progestin therapy. The regression or persistence of endometrial intraepithelial neoplasia or whether it progresses should be assessed with adequate investigations. Although a thorough examination of the uterus after hysterectomy is the ideal method of disease assessment, it is not indicated for patients who opt for conservative treatment. After hormonal therapy for patients with endometrial intraepithelial neoplasia, follow-up surveillance should include a series of endometrial biopsies, scheduled every 3-6 months. However, the most appropriate timing of the examination evaluation has not been clarified.
The choice of the best conservative treatment option for endometrial intraepithelial neoplasia is also inconclusive. Therefore, it is not possible to recommend a standard of care. Table 3 lists several currently recommended therapies, of which oral progestin therapy or placement of levonorgestrel-containing IUDs are preferred. However, the medication should be applied continuously for 12 months without interruption unless the medication is ineffective and the lesion continues to progress. However, the choice of our conservative treatment should depend on the specific clinical condition of the patient. If the patient no longer has fertility requirements or her physical condition improves to allow surgery, our treatment should still prefer surgical removal of the uterus. In many women, abnormal hormonal conditions persist in the body after the end of hormone therapy (resulting in endometrial intraepithelial neoplasia), so when the lesion recurs and it is unclear whether to continue medication, localized lesions should be selected for excision. Since obese women have a high incidence of endometrial cancer and endometrial intraepithelial neoplasia is a precursor to cancer, clinicians may counsel patients to lose weight or opt for bariatric surgery to reduce the chance of recurrence of neoplasia. However, for long-term systemic clinical management of endometrial intraepithelial neoplasia, we should also consider the side effects of treatment – edema, gastrointestinal dysfunction, and thromboembolic events (which are common side effects). Therefore, we should choose a rational treatment decision in the medical management of those patients who are inoperable.