CD10, scored as positive versus negativ Figure 2e 0% 20% 40% 60% 80% 100% BNLI ECA M PO EO RTC G ELA KIEL LUBECK NO RDIC VANCOUVER W URZBURG All NotScored Negative Positive included/excluded "notscored" agreem entin all scores agreementin pairs oflabs generalized kappa firstround excluded 45% 87% 0.69 included 3% 65% 0.39 second round excluded 82% 95% 0.88 included 70% 87% 0.72 all path 1 path 2 path 3 path 4 path 5 path 6 path 7 path 8 path 9 CD10 can be reproducibly scored, but is very sensitive to laboratory variations inducing variation of non- scored cases substantial fair very good substantial
13
Embed
CD10, scored as positive versus negative all path 1 path 2 path 3 path 4 path 5 path 6 path 7 path 8 path 9 CD10 can be reproducibly scored, but is very.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CD10, scored as positive versus negative
Figure 2e
0% 20% 40% 60% 80% 100%
BNLI
ECAMPO
EORTC
GELA
KIEL
LUBECK
NORDIC
VANCOUVER
WURZBURG
All
Not Scored
Negative
Positive
included/excluded "not scored"
agreement in all scores
agreement in pairs of labs
generalized kappa
first round excluded 45% 87% 0.69included 3% 65% 0.39
second round excluded 82% 95% 0.88included 70% 87% 0.72
• Using immunohistochemistry, prognostic biomarkers for DLBCL can be assessed with varying reliability
• Results in almost all markers are strongly influenced by technical variations (MUM-1, Ki-67, bcl-6)
• Immunohistochemical amplification techniques may result in unexpected variations (CD5, bcl-6)
• Results can be reliably interpreted when optimized stains are used and with well-defined scoring guidelines (kappa=0.43-0.88), including definitions of internal controls
• at this stage, risk-stratified treatment of DLBCL should be performed in clinical trials with central pathology review