975P - Preoperative MRI Versus Intra-operative Frozen Section in Surgical Management of Clinically Early Endometrial Cancer

Date 09 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Endometrial Cancer
Gynaecologic Malignancies
Imaging, Diagnosis and Staging
Pathology/Molecular Biology
Presenter Amita Maheshwari
Citation Annals of Oncology (2017) 28 (suppl_5): v330-v354. 10.1093/annonc/mdx372
Authors A. Maheshwari1, R. Ranade2, R. Kerkar3, S. Ts3, K. Deodhar4, B. Rekhi5, S. Menon5, S. Gupta6
  • 1Gynecologic Oncology, TATA MEMORIAL CENTRE, 400012 - MUMBAI/IN
  • 2Gynecologic Oncology, NH Hospital, 400012 - Mumbai/IN
  • 3Gynecologic Ongology, Tata Memorial Centre, 400012 - Mumbai/IN
  • 4Pathology, NH Hospital, 400012 - Mumbai/IN
  • 5Pathology, Tata Memorial Centre, 400012 - Mumbai/IN
  • 6Medical Oncology, Tata Memorial Centre, 400012 - Mumbai/IN

Abstract

Background

The role of systematic lymphadenectomy in clinically early stage endometrial cancer is controversial. A number of factors can predict lymph node metastasis including myometrial invasion, tumor grade in endometrial cancers. The purpose of the present study is to evaluate the accuracy of preoperative MRI and intraoperative frozen section in determining the depth of myometrialinvasion, cervical involvement, tumor size and lymph nodal status. We also studied the accuracy of preoperative endometrial biopsy and intraoperative frozen section in determining the grade of the tumor.

Methods

Medical records of 160 consecutive cases of clinically early stage endometrial cancer were reviewed retrospectively. A record of depth of myometrial invasion, tumor size, cervical involvement and presence of enlarged lymph nodes was made on a preoperative MRI. Similar depth of myometrial invasion, tumor size, cervical involvement and grade of the tumor were recorded on an intraoperative frozen section. The grade of the tumor was also recorded on a preoperative endometrial biopsy. Standard statistical calculations were used.

Results

The sensitivity and specificity of MRI for myometrial invasion were 81.3 and 75%, respectively while that for frozen section were 80 and 96.2%, respectively. For tumor grade the sensitivity and specificity of preoperative endometrial biopsy were 60 and 95.6%, respectively while that of frozen section were 53.8 and 97.6%, respectively. For cervical involvement the sensitivity of MRI and frozen section was 62.5 and 98.4%, respectively.

Conclusions

Although the sensitivity of both frozen section and MRI for predicting deep myometrial invasion was similar (80 vs 81.3%) but the specificity (96.2 vs 75%) and negative predictive value (92.7 vs 88.2%) of frozen section were superior to MRI.Both preoperative biopsy and intraoperative frozen section had low sensitivity (60 vs 53.8%) for detecting a high-grade lesion.

Clinical trial identification

Legal entity responsible for the study

Institutional ethics committee, Tata Memorial Centre

Funding

None

Disclosure

Author has declared no conflicts of interest.