ABSTRACT
Endometrial biopsies are important in the diagnostic workup of women who present with abnormal uterine bleeding or hereditary risk of endometrial cancer. In general, approximately 10% of all endometrial biopsies demonstrate endometrial (pre)malignancy that requires specific treatment. As the diagnostic evaluation of mostly benign cases results in a substantial workload for pathologists, artificial intelligence (AI)-assisted preselection of biopsies could optimize the workflow. This study aimed to assess the feasibility of AI-assisted diagnosis for endometrial biopsies (endometrial Pipelle biopsy computer-aided diagnosis), trained on daily-practice whole-slide images instead of highly selected images. Endometrial biopsies were classified into 6 clinically relevant categories defined as follows: nonrepresentative, normal, nonneoplastic, hyperplasia without atypia, hyperplasia with atypia, and malignant. The agreement among 15 pathologists, within these classifications, was evaluated in 91 endometrial biopsies. Next, an algorithm (trained on a total of 2819 endometrial biopsies) rated the same 91 cases, and we compared its performance using the pathologist's classification as the reference standard. The interrater reliability among pathologists was moderate with a mean Cohen's kappa of 0.51, whereas for a binary classification into benign vs (pre)malignant, the agreement was substantial with a mean Cohen's kappa of 0.66. The AI algorithm performed slightly worse for the 6 categories with a moderate Cohen's kappa of 0.43 but was comparable for the binary classification with a substantial Cohen's kappa of 0.65. AI-assisted diagnosis of endometrial biopsies was demonstrated to be feasible in discriminating between benign and (pre)malignant endometrial tissues, even when trained on unselected cases. Endometrial premalignancies remain challenging for both pathologists and AI algorithms. Future steps to improve reliability of the diagnosis are needed to achieve a more refined AI-assisted diagnostic solution for endometrial biopsies that covers both premalignant and malignant diagnoses.
Subject(s)
Artificial Intelligence , Computers , Humans , Female , Feasibility Studies , Hyperplasia , Reproducibility of Results , BiopsyABSTRACT
BACKGROUND: In endometrial cancer (EC), preoperative anaemia, thrombocytosis and leucocytosis appear to be associated with worse prognosis. It remains unclear whether these parameters solely reflect tumour aggressiveness, or also impact response to adjuvant treatment. Therefore, our primary aim is to evaluate the prognostic relevance of anaemia, thrombocytosis and leucocytosis on survival in EC. Secondary, to explore their predictive relevance in response to radiotherapy in EC. METHODS: A retrospective multicentre cohort study was performed within 10 hospitals. Preoperative haematological parameters were defined as: Anaemia - haemoglobin <7.45 mmol/L (<12 g/Dl), thrombocytosis - platelets >400 × 109 platelets/L, leucocytosis - leukocytes >10 × 109/L. The relationship of haematological parameters with clinicopathological characteristics, ESGO/ESTRO/ESP risk groups and survival were evaluated. Furthermore, the predictive value of haematological parameters was determined on the overall response to adjuvant radiotherapy and for the ESGO/ESTRO/ESP intermediate-risk group solely receiving radiotherapy. RESULTS: A total of 894 patients were included with a median follow-up of 4.5 years. Anaemia was present in 103 (11.5%), thrombocytosis in 79 (8.8%) and leucocytosis in 114 (12.7%) patients. The presence of anaemia or thrombocytosis was significantly associated with ESGO/ESTRO/ESP high-risk (respectively, P = 0.002 and P = 0.041). In the entire cohort, anaemia remained independently associated with decreased disease-specific survival (HR 2.31, 95% CI (1.19-4.50), P = 0.013) after adjusting for age, the abnormal haematological parameters and ESGO/ESTRO/ESP risk groups. In patients that were treated with adjuvant radiotherapy (n = 239), anaemia was associated with significant reduced 5-year disease-specific and recurrence-free survival (P = 0.005 and P = 0.025, respectively). In ESGO/ESTRO/ESP intermediate risk patients that received solely vaginal brachytherapy (n = 74), anaemia was associated with reduced disease-specific survival (P = 0.041). CONCLUSIONS: Current data demonstrate the importance of preoperative anaemia as independent prognostic factor in patients with EC. Moreover, anaemia seems to be associated with reduced response to radiotherapy. Prospective validation in a larger study cohort is needed to verify anaemia as predictive biomarker for radiotherapy.What is already known on this subject? In endometrial cancer, preoperative abnormal haematological parameters like, anaemia, thrombocytosis and leucocytosis appears to be associated with FIGO advanced-stage and unfavourable outcome.What do the results of this study add? It remains unclear whether anaemia, thrombocytosis or leucocytosis solely reflecting worse prognosis by advanced tumour stage, or also impact response to adjuvant treatment. Current data demonstrate that anaemia is independent associated with decreased disease-specific survival and anaemia seems related with reduced response to radiotherapy and in specific to vaginal brachytherapy in ESGO/ESTRO/ESP intermediate risk patients.What are the implications of these findings for clinical practice and/or further research? Specific applied adjuvant treatment is needed if patients with anaemia have a reduced response to radiotherapy in EC. Prospective validation in a larger study cohort is required to verify anaemia as predictive biomarker for radiotherapy and to further evaluate the prognostic/predictive impact of anaemia in addition to the molecular subgroups.
In this study we focused on three specific blood values before surgery to predict survival outcomes in endometrial cancer patients: low haemoglobin (anaemia), high platelet count (thrombocytosis) and high white blood cell count (leucocytosis). We studied 894 patients with endometrial cancer over about 4.5 years, in which 11.5% had anaemia, 8.8% thrombocytosis and 12.7% leucocytosis. Anaemia was linked to a lower chance of surviving endometrial cancer, even after we considering patients' age, thrombocytosis, leucocytosis and the endometrial cancer risk classification groups. In patients who received radiotherapy after surgery (293 patients), anaemia was linked to a lower change of surviving and cancer coming back within 5 years. In patients within the intermediate endometrial cancer risk classification group who only received specific radiotherapy (74 patients), anaemia was even linked with lower chance of survival. In conclusion, anaemia is an important factor in predicting endometrial cancer outcomes, and it might also make radiotherapy less effective for some patients.
Subject(s)
Anemia , Endometrial Neoplasms , Thrombocytosis , Female , Humans , Anemia/etiology , Biomarkers , Cohort Studies , Endometrial Neoplasms/complications , Endometrial Neoplasms/radiotherapy , Endometrial Neoplasms/surgery , Leukocytosis , Thrombocytosis/etiology , Retrospective StudiesABSTRACT
Prehabilitation is an upcoming strategy to optimize patient's functional capacity, nutritional status, and psychosocial well-being in order to reduce surgical complications and enhance recovery. This study aims to assess the feasibility of implementing a multimodal prehabilitation program into the standard care of gynecological oncology patients at an academic hospital in terms of recruitment, adherence, and safety, which were assessed by the number of patients eligible, recruitment rate, participation rate, and adherence to individual modalities. Data were derived from the F4S PREHAB trial, a single-center stepped-wedge trial implementing a multimodal prehabilitation program among various surgical specialties. All patients undergoing elective surgery as part of treatment for ovarian, uterine, and vulvar cancer at the Radboudumc, an academic hospital in The Netherlands, between May 2022 and September 2023 were considered eligible for the F4S PREHAB trial and, consequently, were included in this cohort study. The multimodal prehabilitation program comprised a physical exercise intervention, nutritional intervention, psychological intervention, and an intoxication cessation program. A total of 152 patients were eligible and approached for participation of which 111 consented to participate, resulting in a recruitment rate of 73%. Participants attended an average of six exercise sessions and adhered to 85% of possible training sessions. Respectively, 93% and 98% of participants adhered to the prescribed daily protein and vitamin suppletion. Ten participants were referred to a psychologist and completed consultations. Out of nine active smokers, two managed to quit smoking. A total of 59% adhered to alcohol cessation advice. No adverse events were reported. This study demonstrates that introducing a multimodal prehabilitation program into the standard care of gynecological oncology patients is feasible in terms of recruitment and adherence, with no serious adverse events.
ABSTRACT
BACKGROUND: In the last decade, technical innovations have resulted in the development of several minimally invasive diagnostic cancer tools. Within women at high risk of developing ovarian or endometrial cancer (EC) due to hereditary cancer syndrome, there is an urgent need for minimally invasive and patient-friendly methods to detect ovarian cancer and EC at an early stage. MATERIALS AND METHODS: We performed a systematic search of studies using DNA methylation or mutation analysis, microbiome, or proteomics performed on cervicovaginal specimens (smear, swab, or tampon) intended to detect ovarian and EC published until January 2024. RESULTS: Included studies (n = 36) showed high heterogeneity in terms of biomarkers used and outcomes, and only a few studies reported on the detection of biomarkers in high-risk subgroups. CONCLUSION: Based on the findings in this review, DNA methylation techniques seem to be the most promising for detecting ovarian and EC at early stages in the general population. Future validation of cervicovaginal DNA methylation techniques is needed to determine whether this technique might be beneficial in hereditary high-risk subgroups.
Subject(s)
Biomarkers, Tumor , DNA Methylation , Early Detection of Cancer , Endometrial Neoplasms , Ovarian Neoplasms , Humans , Female , Endometrial Neoplasms/diagnosis , Endometrial Neoplasms/genetics , Endometrial Neoplasms/metabolism , Early Detection of Cancer/methods , Ovarian Neoplasms/diagnosis , Ovarian Neoplasms/genetics , Biomarkers, Tumor/genetics , Vagina/metabolism , Vagina/microbiology , Cervix Uteri/metabolism , Cervix Uteri/pathology , Proteomics/methodsABSTRACT
BACKGROUND: Response to hormonal therapy in advanced and recurrent endometrial cancer (EC) can be predicted by oestrogen and progesterone receptor immunohistochemical (ER/PR-IHC) expression, with response rates of 60% in PR-IHC > 50% cases. ER/PR-IHC can vary by tumour location and is frequently lost with tumour progression. Therefore, we explored the relationship between ER/PR-IHC expression and tumour location in EC. METHODS: Pre-treatment tumour biopsies from 6 different sites of 80 cases treated with hormonal therapy were analysed for ER/PR-IHC expression and classified into categories 0-10%, 10-50%, and >50%. The ER pathway activity score (ERPAS) was determined based on mRNA levels of ER-related target genes, reflecting the actual activity of the ER receptor. RESULTS: There was a trend towards lower PR-IHC (33% had PR > 50%) and ERPAS (27% had ERPAS > 15) in lymphogenic metastases compared to other locations (p = 0.074). Hematogenous and intra-abdominal metastases appeared to have high ER/PR-IHC and ERPAS (85% and 89% ER-IHC > 50%; 64% and 78% PR-IHC > 50%; 60% and 71% ERPAS > 15, not significant). Tumour grade and previous radiotherapy did not affect ER/PR-IHC or ERPAS. CONCLUSIONS: A trend towards lower PR-IHC and ERPAS was observed in lymphogenic sites. Verification in larger cohorts is needed to confirm these findings, which may have implications for the use of hormonal therapy in the future.