Optimal patient selection for personalized treatment in Georgian Gastric Cancer population
Session type: E-poster/poster
Gastric cancer (GC) is one of the most common cancers worldwide, In our study we performed an innovative pathological assessment through scoring of individual biopsies against whole biopsies from single patients. Herein, we also employed statistical methods to estimate the false negative detection risks when analyzing finite numbers of biopsies in order to understand the relationship between the number of biopsies and the risk of selecting a false positive patient for a particular treatment or inclusion in a clinical trial.
From 2012-2017 in different centers were collected 112 patients with multiple biopsies from different tumor areas. In Lab of TSMU were performed Immunohistochemistry (IHC) Fluorescence in situ hybridization (FISH). biomarker status including IHC and FISH staining of MET, ATM, FGFR2 and HER2 were evaluated on each biopsy. Individual scores for each biomarker were given to each biopsy. For MET IHC, MET FISH, FGFR2 FISH and HER2.
The positive biopsy numbers were slightly less than the total biopsy numbers. In the 11 MET IHC positive cases, 64% of the cases showed low heterogeneity, while 36% showed medium and 3% showed high heterogeneity, significant correlation was found between MET IHC score and MET FISH results (p<0.01,κ=0.58, Fisher’s exact test) In the 29 cases with at least one ATM negative biopsyIn the 5 FGFR2 FISH positive cases, 56% of the cases showed low heterogeneity, while 33% showed medium and11% showed high heterogeneitIn the 16 HER2 positive cases, 77% of the cases showed low heterogeneity, while 23% showed medium heterogeneity and none of the cases showed high heterogeneity.
Our results show a relatively low level of heterogeneity across the biomarkers analyzed in this cohort, a decrease in the rate of false negative detection corresponding with an increase in the biopsy number for all biomarkers tested herein, demonstrating the benefit of multiple biopsy sampling and serving as an example of addressing intra-tumoral heterogeneity using statistical methods.
In our study, in order to better evaluate intratumoral heterogeneity, we employed surgical biopsy as our tumor sampling strategy, we performed an innovative pathological assessment through scoring of individual biopsies against whole biopsies from single patients.