Virtual Library
Start Your Search
Shireen Vali
Author of
-
+
P1.04 - Immunooncology (Not CME Accredited Session) (ID 936)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
-
+
P1.04-20 - Computational Biological Model Prediction of PD-L1 Expression and Immunotherapy Response for KRAS Mutated Lung Cancer Based on Co-Mutations (ID 13049)
16:45 - 18:00 | Author(s): Shireen Vali
- Abstract
Background
Emerging data suggest that KRAS mutated non-small cell lung cancer (NSCLC) is a heterogeneous disease based on the presence of co-mutations. These co-mutations may impact PD-L1 expression, a predictive biomarker for PD-1/PD-L1 immunotherapy, and may result in differential responses to immunotherapy.
a9ded1e5ce5d75814730bb4caaf49419 Method
Genomic information of NSCLC patients, including 2888 from publically available datasets and 86 from Stanford University, was input into computational biological model (CBM) software (Cellworks Group, San Jose, CA). Customized computational protein network maps of disease characteristics were generated for each patient. CBM was used to predict PD-L1 protein expression and also response to PD-1/PD-L1 immunotherapy in KRAS co-mutation subsets using 3 key metrics: PD-L1 expression; Dendritic Cell Infiltration Index (9 chemokine markers); and Immunosuppressive Biomarker Expression (14 markers).
4c3880bb027f159e801041b1021e88e8 Result
The major co-mutations observed with the KRAS mutation were in tumor suppressor genes (TP53, STK11, CDKN2A, KEAP1) and a downstream effector (PIK3CA). Using the CBM approach, KRAS mutated NSCLC tumors with TP53 co-mutations had the highest prevalence of PD-L1 protein expression whereas tumors with KRAS/KEAP1 and KRAS/STK11/KEAP1 co-mutations were associated with the lowest expression. Expression of PD-L1 in tumors with KRAS/STK11, KRAS/CDKN2A, KRAS/PIK3CA co-mutations, and KRAS without co-mutations was higher than in tumors with KRAS/STK11/KEAP1 and KRAS/KEAP1 co-mutations. Of the 30 NSCLC tumors in the Stanford dataset with available PD-L1 immunohistochemistry results, including 19 with KRAS/TP53 and 11 with KRAS/KEAP1 or KRAS/STK11/KEAP1, CBM accurately predicted PD-L1 expression in these two groups at rates of 79% and 72%, respectively. In regards to prediction of response to PD-1/PD-L1 immunotherapy, CBM predicted the majority of patients with KRAS/KEAP1 and KRAS/STK11/KEAP1 to be non-responders, whereas CBM predicted the majority of patients with KRAS/TP53, KRAS/PI3KCA, and KRAS without co-mutations to be responders. The proposed mechanism for KRAS co-mutations’ impact on PD-L1 expression from the CBM model integrates differential activation of (i) downstream pathways of KRAS (PI3K/AKT, RAF/ERK, and RAL) and (ii) transcription factors involved in PD-L1 expression (i.e., MYC, HIF1α, NFKβ, AP1, and STAT1/3).
8eea62084ca7e541d918e823422bd82e Conclusion
KRAS mutated NSCLC is emerging as a diverse disease based on co-mutations. The CBM approach demonstrates that PD-L1 expression varies among KRAS co-mutation subtypes along with likelihood of response to PD-1/PD-L1 immunotherapy. CBM provides proposed mechanisms underlying these differences and therefore, provides further rationale to examine more precise delivery of immunotherapy.
6f8b794f3246b0c1e1780bb4d4d5dc53
-
+
P1.14 - Thymoma/Other Thoracic Malignancies (Not CME Accredited Session) (ID 946)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
-
+
P1.14-17 - Identification of Molecular Subtypes of Thymic Epithelial Tumors and Novel Treatments Using a Computational Biological Model (ID 12521)
16:45 - 18:00 | Author(s): Shireen Vali
- Abstract
Background
The histologic classification of thymic epithelial tumors (TETs) is based on the description of both epithelial cell morphology and relative abundance of lymphocytes. Here, we used a computational biological model (CBM) approach on The Cancer Genome Atlas (TCGA) dataset to identify molecular subtypes of TETs and associated predicted therapeutic options.
a9ded1e5ce5d75814730bb4caaf49419 Method
Whole exome sequencing and gene expression data from the TCGA TET dataset (n = 102) along with the IUTAB-1 cell line was input into CBM software (Cellworks Group, San Jose, CA) to build an unsupervised classification model beyond molecular subtypes previously reported (Loehrer PJ ASCO 2017). The CBM generated a disease specific protein network map using PubMed and other online resources. Using computer simulation, disease biomarkers unique to each tumor were identified within the protein network maps. Among the tumors simulated, 6 molecular clusters were identified (TH1-TH6). The CBM digital drug library was tested against these molecular subtypes and the cell growth score (i.e. cell proliferation, viability, and apoptosis) was analyzed.
4c3880bb027f159e801041b1021e88e8 Result
The CBM identified 6 molecular subtypes among 102 TET patients. Among subtypes with a GTF2I mutation, TH1, TH4, and TH6 also had chromosomal aberrations in chromosome 22 and 9. Deletion of chromosome 22 was present in TH1, deletion of chromosome 9 in TH4 and TH6, and also amplification of chromosome 22q in TH4. Among GTF2I wild type subtypes, chromosome 22q deletion and complex cytogenetics were present in TH2, trisomy of chromosome 1 in TH3, and HRAS mutations and chromosome 2 amplification in TH5. The IUTAB-1 cell line had a GTF2I mutation and mapped to the TH4 molecular subtype. The CBM predictions of sensitivity of TH4 subtype to Nelfinavir (AKT inhibitor) and Panobinostat (histone deacetylase inhibitor) along with resistance to Everolimus (MTOR inhibitor) were validated in vitro. There were two molecular subtypes for which Everolimus was predicted to be sensitive, TH1 and TH6.
8eea62084ca7e541d918e823422bd82e Conclusion
We present an updated classification of TETs based on a CBM approach and associated potential novel therapeutic options that could be further validated in clinical trials.
6f8b794f3246b0c1e1780bb4d4d5dc53
