Abstract 96P
Background
In this study, we report the impact of CGP and resulting personalized therapy (tx) selection on cancer patient (pt) survival.
Methods
We conducted retrospective medical record review of 194 pts with solid tumors, age ≥ 18 years who underwent CGP Jan 2016-Dec 2020. Molecular Tumor Board (MTB) case discussions began Feb 2018. We reviewed all genomic alterations, potential targeted tx options and the impact of molecular guided tx selection on survival. Cox Proportional Hazard model for survival analysis was adjusted for cancer stage, age, time of testing, and duration of targeted tx.
Results
194 pts had 243 total CGP tests performed (1.25 tests/pt) including 191 tissue samples and 52 liquid biopsies. Of 154 eligible pts, 56 (36%) were discussed at the MTB. The most frequent potentially targetable genes in the entire cohort were PIK3CA (15%), EGFR (8%), MTAP (6%), ERBB2 (6%), NF1 (5%), BRCA2 (4%), FGFR1 (4%), MET (4%), NOTCH1 (4%) and BRCA1 (3%). The most frequent untargetable alterations were in TP53 (51%), CDKN2A/B (16%/13%), TERT (14%), and KRAS non G12C (13%). We assessed the impact of covariates on survival by dividing subjects into a group with genomic alternations (GA) that have a matched targeted tx (Group A, 129 pts) and a group without targetable GA (Group B, 65 pts). Group A subjects were treated with targeted tx matched to CGP results. Group B subjects did not receive targeted tx. Mean duration of targeted tx employed in group A was 3.23 months. Median overall survival (OS) in group A with targeted tx was 810 days vs. 750 days without targeted tx (p=0.0056). Cox regression analysis revealed the proportional relation between duration of treatment and decreased risk of death (aHR=0.9589, p= 0.00742). Also, pts whose primary tumors had not metastasized at time of CGP with actionable GA had decreased risk of death compared to pts with no actionable GA (aHR 0.38, P=0.015).
Conclusions
Despite small sample size, retrospective nature, physician bias in requesting CGP and the availability of targeted tx, this study from a single institution confirms that genomic profiling of pts with clinically advanced solid tumors identifies targeted tx opportunities that result in improved overall pt survival.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Foundation Medicine Inc.
Disclosure
All authors have declared no conflicts of interest.