1Michael J Pishvaian, 2Edik M Blais, 3Lynn M Matrisian, 4Jonathan R Brody, 2Dzung Thach, 2Patricia DeArbeloa, 4Flavio G Rocha, 5Jennifer W Chuy, 6Raymond C Wadlow, 7Andrew E Hendifar, 8,2Emanuel F Petricoin III
1 Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Washington, DC; 2 Perthera, Mclean, VA; 3 Pancreatic Cancer Action Network, El Segundo, CA; 4 Oregon Health & Science University, Portland, OR; 5 NYU Langone, New York, NY; 6 Inova Schar Cancer Institute, Fairfax, VA; 7 Cedars-Sinai Medical Center, Los Angeles, CA; 8 George Mason University, Manassas, VA
Content Type: Poster
BACKGROUND:
- Nearly 50% of pts with mPDAC never receive a 2nd line therapy for metastatic disease following frontline FFX or GA.
- Genomic alterations in the DDR pathway2 (e.g. BRCA1/2) are associated with longer progression-free survival (PFS) on platinum-containing regimens (e.g. FFX). PDACai uses the full NGS profile to predict benefit from GA and/or FFX in mPDAC.
- PDACai v2.0 redefines the mutational landscape in mPDAC by removing ATM mutations from the DDR pathway (top FFX feature) and adding new variant-level features (e.g. TP53 GOF).
METHODS:
PDACai learned to predict outcomes based on NGS testing results from Training Cohorts where pts received either FFX & GA in the 1st line setting. PDACai predictions were independently analyzed for FFX & GA Validation Cohorts.
PDACai predictions were analyzed retrospectively for pts who received a Perthera Report. Outcomes were captured via the Perthera Registry. US pts whose physicians order a Perthera Report may be eligible to participate in prospective efforts to validate PDACai (frontline/adjuvant, non-interventional).
RESULTS:
PDACai predictions were normalized and split equally into thirds (Upper, Middle, Lower) for each therapy option (FOLFIRINOX vs gemcitabine/nab-paclitaxel).
1st Line FOLFIRINOX (FFX) progression-free survival (PFS) and overall survival (OS) outcomes stratified by PDACai were significantly different using an independent validation cohort (n=165).
-Upper: PFS = 14.1 months, OS = 2.2 years
-Middle: PFS = 10.1 months, OS = 1.5 years
-Lower: PFS = 9.2 months, OS = 1.7 years
1st Line gemcitabine/nab-paclitaxel (GA) PFS and OS outcomes stratified by PDACai were also significantly different using an independent validation cohort (n=148)
-Upper: PFS = 8.8 months, OS = 1.9 years
-Middle: PFS = 6.5 months, OS = 1.4 years
-Lower: PFS = 5.3 months, OS = 1.3 years
View a PDF of the poster to see KM Curves
CONCLUSIONS:
- Response to chemotherapy is heterogeneous and difficult to predict in pts with mPDAC
- PDACai v2.0 successfully predicted differences in PFS for both FFX and GA cohorts (better than v1.0)
- Guidance for 1st line decisions is now feasible with upfront NGS and PDACai v2.0 (faster than v1.0)
- Efforts to validate PDACai predictions for 2nd line variations of FFX (FOLFIRI vs FOLFOX) are ongoing
- Continuous prospective validation of AI/ML models that utilize lab-agnostic NGS results is warranted