University Medical Center endometrial sampler (Tao brush) with a new, added built-in suction component was evaluated in this ex-vivo study. Materials and ...
ENDOMETRIAL BIOPSY USING A NOVEL ENDOMETRIAL SAMPLER IS A RELIABLE METHOD FOR DETECTING ENDOMETRIAL CANCER WITH HIGH INTEROBSERVER AGREEMENT 1Howard H. Wu, MD; 1Robert E. Emerson, MD; 1Shaoxiong Chen, MD, PhD; 3M. Joe Ma, MD, PhD; 1Muhammad Idrees, MD, 1Xiaoyan Wang, MD, PhD; 1Harvey M Cramer, MD, 2Giuseppe Del Priore, MD, MPH 1Indiana University School of Medicine, Department of Pathology and Laboratory Medicine and IU Health Pathology Laboratory, Indianapolis, IN 2Cancer Treatment Centers of America, Schaumburg IL 3Florida Hospital Orlando, Department of Pathology, Orlando FL
ABSTRACT
MATERIALS AND METHODS Table I Diagnoses of GDP-Tao endometrial sampling among 6 pathologists with comparison to the final hysterectomy diagnoses
Introduction: A novel outpatient endometrial sampler (GDP-Tao, Cook Medical) that is a modification of the original Indiana University Medical Center endometrial sampler (Tao brush) with a new, added built-in suction component was evaluated in this ex-vivo study. Materials and Methods: The GDP-Tao sampler was used to obtain endometrial cells and tissues from 35 fresh hysterectomy specimens. If there was scant tissue, only one SurePath slide was made (6 cases). Otherwise, both H&E-stained cell block sections and SurePath slides were prepared (29 cases). Six pathologists who were blinded to the final hysterectomy diagnoses interpreted the GDP-Tao biopsies independently, using a set of diagnostic terms including non-diagnostic (ND), benign (B), atypia of unknown significance (AUS), atypical hyperplasia (AH), suspicious for malignancy (SM) and malignant (M). The GDP-Tao diagnoses were correlated with the final diagnosis established by the hysterectomy specimens. Results: 17 out of 18 cases of AH/M were correctly diagnosed by the GDP-Tao. 11/11 samples were benign on both GDP-Tao and final pathology. One case diagnosed as benign by GDP-Tao showed focal AH in the hysterectomy specimen and one GDP-Tao case diagnosed as AUS showed only benign findings on the correlating hysterectomy. Among 4 ND samples on GDP-Tao, three hysterectomies showed benign endometrium and one lacking endometrium from recent ablation. The overall sensitivity, specificity, positive predictive and negative predictive values were 95%, 92%, 95% and 92%, respectively. There is a very high level of interobserver agreement (Cronbach’s alpha = 0.988, p< 0.001) with only three cases in which the GDP-Tao diagnoses demonstrated disagreement between six pathologists. Conclusions: Pathologic interpretations using a combined cytologic and histologic cytologic approach on ex-vivo samples obtained by the new, outpatient GDP-Tao endometrial sampler provides a highly sensitive and specific method for accurately detecting endometrial cancer and can be performed with a high degree of interobserver agreement.
Number
Age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
35 41 67 64 53 43 41 78 57 44 67 83 52 55 50 29 52 64 47 51 57 68 43 86 70 51 49 34 77 64 77 69 47 69 63
CB Yes Yes Yes Yes Yes Yes Yes No No Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No Yes Yes Yes Yes Yes Yes Yes Yes
SurePath No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Pathologist 1 B B S M M M ND M A B AH M B M B ND B M B M AH M B M ND ND B B B A M M B M M
Figure 1 GDP-Tao endometrial device
Pathologist 2 B B M M M M ND M S B M M B AH ND ND B AH B M M M B AH ND ND B B B S M AH B M M
Pathologist 3 B B M M M M ND M B B M M B M B B B M B M M M B M ND B B B B B M M B M M
Pathologist 4 B B S M M M ND M AUS B M M B M B ND B M B M M M B S ND ND B B B B M M B M M
Pathologist 5 B B AUS M AUS M ND M AUS B AH M B M AUS ND B M B M AH M B AUS ND ND ND B B ND M M B M M
Pathologist 6 B B S M M M ND M AUS B AH M B M B ND B AH B M AH M B S ND ND ND B B AUS M S B M M
Figure 2A Proliferative endometrium, SurePath cytology, Papanicolaou-stained x100
FINAL B B M M M M ND M B B AH/M AH/M B M B B B M B M M M AH M B B B B B AH M M B M M
Final hysterectomy Dx Proliferative em Proliferative em Serous ca in situ Serous carcinoma endometriod adenca gr I PD carcinoma absent endometrium s/p ablation serous carcinoma inactive endometrium, polyp Proliferative em atypical hyperplasia with WD adenocarcinoma atypical hyperplasia with WD adenocarcinoma Weakly proliferative em endometriod adenca gr II secretory em Proliferative em Atrophic endometrium WD endometrioid adenocarcinoma Proliferative em WD endometrioid adenocarcinoma WD endometrioid adenocarcinoma Malignant mixed mullerian tumor complex hyperplasia with atypia Endometrioid adenocarcinoma Inactive endometrium Atrophic endometrium Atrophic endometrium Proliferative em Atrophic endometrium Adenomyosis with CAH Endometriod adenocarcinoma Endometriod adenocarcinoma gr I Benign basalis endometrium Mixed endometrioid and clear cell ca Endometrioid adenocarcinoma, gr 2
Figure 2B Proliferative endometrium, cell block, H&E-stained x100
RESULTS 17 out of 18 cases of AH/M were correctly diagnosed by the GDP-Tao. 11/11 samples were benign on both GDP-Tao and final pathology. One case diagnosed as benign by GDP-Tao showed focal AH in the hysterectomy specimen and one GDP-Tao case diagnosed as AUS showed only benign findings on the correlating hysterectomy. Among 4 ND samples on GDP-Tao, three hysterectomies showed benign endometrium and one lacking endometrium from recent ablation. The overall sensitivity, specificity, positive predictive and negative predictive values were 95%, 92%, 95% and 92%, respectively. There is a very high level of interobserver agreement (Cronbach’s alpha = 0.988, p< 0.001) with only three cases (case 9, 15 and 30) in which the GDP-Tao diagnoses demonstrated disagreement between six pathologists. In 80% of cases the diagnoses were made based on histology alone, 15% of cases based on cytology alone and 5% of cases depend on both cytology and histology.
CONCLUSIONS • Pathologic interpretations using a combined cytologic and histologic approach on ex-vivo samples obtained by the new, outpatient GDP-Tao endometrial sampler provides a highly sensitive (95%) and specific (92%) method for accurately detecting endometrial cancer. • With the preparation of H&E stained histology sections and complimentary SurePath cytology, there is a high degree of interobserver agreement for the diagnosis among pathologists who practice in both community and academic setting.
INTRODUCTION We investigated a new device to evaluate a more effective way of collecting endometrial samples. This study evaluated a new device which combines Global endometrial Disruption using a brush with a built in suction Process (GDP-Tao). This modified aspiration and disrupting device (GDP-Tao) draws a vacuum and aspirates the tissue sample into the device sheath similar to other vacuum biopsy sampling devices (Figure 1). We compared endometrial biopsy specimens obtained using the GDP-Tao and results from final pathology of the hysterectomy specimens and also the interobserver agreement among 6 pathologists.
The GDP-Tao sampler was used to obtain endometrial cells and tissues from 35 fresh hysterectomy specimens. If there was scant tissue, only one SurePath slide was made (6 cases). Otherwise, both H&E-stained cell block sections and SurePath slides were prepared (29 cases). Six pathologists who were blinded to the final hysterectomy diagnoses interpreted the GDP-Tao biopsies independently, using a set of diagnostic terms including non-diagnostic (ND), benign (B), atypia of unknown significance (AUS), atypical hyperplasia (AH), suspicious for malignancy (SM) and malignant (M). More specific diagnoses such as proliferative endometrium (Figure 2), secretory endometrium, atrophic endometrium, hyperplasia without atypia, complex atypical hyperplasia (Figure 3), well-differentiated endometrioid adenocarcinoma (Figure 4), poorly differentiated adenocarcinoma and serous carcinoma (Figure 5) were also rendered. The GDP-Tao diagnoses were correlated with the final diagnosis established by the hysterectomy specimens. The inter-rater agreement was measured using the Cronbach’s alpha (SPSS v.21 2012 IBM Statistics).
REFERENCES Figure 3A Serous carcinoma, SurePath Figure 3B Serous carcinoma, cell block, cytology, Papanicolaou-stained x400 H&E-stained x400
Figure 4 Complex atypical endometrial hyperplasia, cell block, H&E-stained x 200
Figure 5 Well-differentiated endometrioid adenocarcinoma, cell block, H&E-stained x 200
1. Wu HH, Harshbarger KE, Berner HW, Elsheikh TM. Endometrial brush biopsy (Tao brush): histologic diagnosis of 200 cases with complimentary cytology- an accurate sampling technique in the detection of endometrial abnormalities. Am J Clin Pathol, 2000; 114:412-418
