pathologists, as part of the UNITED study Marloes Smit, PhD-student - - PowerPoint PPT Presentation

Marloes Smit, PhD-student

ALL AUTHORS DECLARE NO CONFLICT OF INTEREST

E-learning for instruction of scoring the tumour-stroma ratio (TSR) in colon carcinoma amongst international pathologists, as part of the UNITED study

The tumour-stroma ratio (TSR) parameter is based on the amount of stroma within the primary tumour. A high stroma percentage predicts for worse disease outcome. The TSR can be determined at routine pathology. The TSR is validated in various (international) studies.

Background

2 UNITED study – www.watchstroma.com 15-Oct-19

Tumour-stroma ratio

3 UNITED study – www.watchstroma.com 15-Oct-19

Stroma-high tumour Stroma-low tumour

Survival

4 UNITED study – www.watchstroma.com 15-Oct-19 5-years OS Stroma-low: 78% Stroma-high: 18% 5-years DFS Stroma-low: 69% Stroma-high: 12%

Scoring TSR

5 UNITED study – www.watchstroma.com 15-Oct-19

View with a 2.5x or 5x

• bjective and find the

part with the most stroma visible

Scoring TSR

6 UNITED study – www.watchstroma.com 15-Oct-19

Zoom in and view that area with a 10x

• bjective.

Scoring TSR

7 UNITED study – www.watchstroma.com 15-Oct-19

A scoring area should contain:

• Tumour cells on four sides
• As much as stroma possible
• No muscle or necrosis

Estimate the percentage for the amount of stroma, in increments of 10%.

The UNITED study → Uniform Noting for International application of the Tumour-stroma ratio as Easy Diagnostic tool The study is designed to prepare implementation of the TSR in routine diagnostic pathology

Methods (1)

8 UNITED study – www.watchstroma.com 15-Oct-19

The UNITED study consists of 4 elements: a. Training national and international pathologists using an E-learning b. Validation in large prospective cohort in international setting c. Automation of the TSR d. Prepare implementation for shared decision making

Methods (2)

9 UNITED study – www.watchstroma.com 15-Oct-19

Results (1)

10 UNITED study – www.watchstroma.com 15-Oct-19

Results (2)

11 UNITED study – www.watchstroma.com 15-Oct-19

Training set

• No. of participants finished 36

Median kappa value 0.72 Minimum kappa value 0.21 Maximum kappa value 0.90

Results (3)

12 UNITED study – www.watchstroma.com 15-Oct-19

Test set

• No. of participants finished 34

Median kappa value 0.77 Minimum kappa value 0.51 Maximum kappa value 0.97

Results (4)

13 UNITED study – www.watchstroma.com 15-Oct-19

Repetition set

• No. of participants finished

31 Median kappa value 0.76 Minimum kappa value 0.60 Maximum kappa value 0.89

Results (5)

14 UNITED study – www.watchstroma.com 15-Oct-19

Intra-observer

• No. of participants finished

31 Median kappa value 0.77 Minimum kappa value 0.61 Maximum kappa value 0.95

Significant improvement from training to test set: p=0.002 No change from test to repetition test set: p=0.299

Results (6)

15 UNITED study – www.watchstroma.com 15-Oct-19

TSR can be scored during diagnostics in 1-2 minutes on H&E stained sections ~ 70% passed the training set only by auto-instruction ~ 90% passed the test set by training No relapse after 2 months wash-out period E-learning is an effective method to instruct pathologists and pathology trainees for scoring TSR, confirming the reproducibility of the TSR scoring method.

Conclusion

16 UNITED study – www.watchstroma.com 15-Oct-19

Collaborations and funding

17 UNITED study – www.watchstroma.com 15-Oct-19

Contact information Email: UNITED@lumc.nl Website: www.watchstroma.com Participating pathologists and residents Other collaborations

• European Society of Pathology (ESP)
• Els Dequeker, Leuven University, EQA
• Jeroen vd Laak, Radboud University Medical

Center

Department of Surgery LUMC

• Marloes Smit
• Gabi van Pelt
• Wilma Mesker
• Rob Tollenaar

Datacenter LUMC

• Annet Roodvoets
• Elma van Meershoek