Abstract 1139P
Background
CUP accounts for 3-5% of all malignant epithelial tumors. It is a heterogeneous group of cancers with a poor prognosis and comprehensive genomic profiling may provide therapeutic insight.
Methods
The molecular profiles of CUP tumors within the CARIS Life Sciences database were analyzed utilizing CODEai, a platform that integrates real-world clinical information obtained from insurance claims and medical records with genomic data. Overall survival (OS) in this real-world dataset was assessed by Kaplan-Meier estimates.
Results
This real-world cohort consists of 3841 tumors noted as CUP: 2137 as adenocarcinoma (ADC), 385 as squamous cell carcinoma (SQ), and 1319 as carcinoma not otherwise specified (NOS). NOS was excluded in histology comparisons but grouped with ADC and SQ as CUP-ALL for further analyses. CUP-SQ had a median OS (mOS) of 559 days (d) vs 265d in CUP-ADC (HR: 0.59, 95% CI [0.51-0.69], p <0.00001). CUP-ALL treated with immunotherapy had a mOS of 601d vs 372d in the CUP-ALL treated with chemotherapy (p<0.01). CUP-ALL treated with targeted therapy had a mOS of 638d vs 374d in the CUP-ALL treated with chemotherapy (p<0.01). Within CUP-ALL, TMB high (HR=0.637, [0.515-0.787], p<0.0001), KRAS WT (HR=0.711, [0.62-0.818], p<0.001), and MSI high tumors (HR=0.54, [0.34-0.85], p=0.006) had a longer mOS vs TMB low, KRAS mutant, and MSI stable counterparts, respectively; no difference was seen with TP53 mutation (HR=0.937, p=0.3). In CUP-ADC, the effects of TMB-H and KRAS WT were not seen, but TP53 WT had a longer mOS of 305d vs 232d with TP53 mutant (HR=0.849, [0.724-0.995], p=0.043). In CUP-ALL and CUP-ADC, no survival difference was seen for PD-L1 positive vs. PD-L1 negative; however, in CUP-SQ, PD-L1 positive had a longer mOS of 769d vs 508d with PD-L1 negative (HR=0.67, [0.46-0.97], p=0.033). In CUP-SQ, TMB-H also showed a longer mOS of 761d vs 382d in TMB-L (HR=0.487, [0.274-0.87], p=0.012).
Conclusions
Our data from a large real-world cohort demonstrates that key molecular alterations have prognostic and predictive roles in CUP. To maximize clinical benefit, prospective studies with various therapeutic classes of cancer treatments exploiting these differences are warranted.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
University of Cincinnati.
Funding
Has not received any funding.
Disclosure
D. Georlette: Financial Interests, Institutional, Sponsor/Funding: Caris Life Sciences. W.M. Korn: Financial Interests, Personal and Institutional, Full or part-time Employment: Caris Life Sciences; Financial Interests, Institutional, Advisory Role: Merck. J. Xiu: Financial Interests, Institutional, Sponsor/Funding: Caris Life Sciences. H. Babiker: Financial Interests, Institutional, Advisory Board: Caris; Financial Interests, Institutional, Advisory Board: Idera; Financial Interests, Institutional, Advisory Board: Myovant; Financial Interests, Institutional, Speaker’s Bureau: Guardant360. P.M. Coelho Barata: Financial Interests, Institutional, Speaker’s Bureau: Caris Life Sciences; Financial Interests, Institutional, Speaker’s Bureau: Bayer; Financial Interests, Institutional, Research Grant: BlueEarth Diagnostics; Financial Interests, Institutional, Other, Contracted Research: AstraZeneca; Financial Interests, Institutional, Advisory Role: Astellas; Financial Interests, Institutional, Advisory Role: Eisai; Financial Interests, Institutional, Advisory Role: EMD Serono; Financial Interests, Institutional, Advisory Role: Dendreon; Financial Interests, Institutional, Advisory Role: Pfizer; Financial Interests, Institutional, Advisory Role: Seattle Genetics. D. Sohal: Financial Interests, Personal, Advisory Role: Ability Pharmaceuticals; Non-Financial Interests, Personal, Other, Honoraria: Foundation Medicine; Financial Interests, Personal, Speaker’s Bureau: Incyte; Financial Interests, Institutional, Research Grant: Amgen; Financial Interests, Institutional, Research Grant: Apexigen; Financial Interests, Institutional, Research Grant: Bristol-Myers Squibb; Financial Interests, Institutional, Research Grant: Celgene; Financial Interests, Institutional, Research Grant: FibroGen; Financial Interests, Institutional, Research Grant: Genentech; Financial Interests, Institutional, Research Grant: MedImmune; Financial Interests, Institutional, Research Grant: Merck; Financial Interests, Institutional, Research Grant: OncoMed; Financial Interests, Institutional, Research Grant: Rafael. All other authors have declared no conflicts of interest.