Abstract C023: PDAC tumors from patients treated with combination MEK and autophagy inhibition reveal a treatment sensitive epithelial subtype that is enriched in murine models
Brian W Labadie,Aleksandar Z Obradovic,Sarah Sta Ana,Michael May,Naomi Sender,Isabelle V Ross,Monica Rodriguez-Mendez,Urszula Wasko,Lorenzo Tomassoni,Li Li,Manan Vij,Sinan Abuzaid,Julia Wulfkuhle,Winston Wong,Ilenia Pellicciotta,Benjamin Izar,Emanuel F Petricoin,Aik Choon Tan,Kenneth P Olive,Alex G Raufi,Gulam A Manji
DOI: https://doi.org/10.1158/1538-7445.pancreatic24-c023
IF: 11.2
2024-09-17
Cancer Research
Abstract:Combining MEK and autophagy inhibitors in preclinical PDAC mouse models resulted in tumor regressions and improved rodent survival. These findings provided the rationale for clinical investigation of this combination. The phase 1 MEKiAUTO trial evaluated combination cobimetinib, HCQ, with or without atezolizumab in 14 patients (pts) with KRAS-mutated pancreatic adenocarcinoma (PDAC). The study was closed due to a lack of preliminary efficacy and poor tolerability; however, three patients experienced progression free survival (PFS) beyond 12 weeks (two with KRAS G12R , one with KRAS G12D ). Correlative analyses on twenty-one biopsies (8-paired FFPE and 6-paired fresh frozen specimens) with multiplex immunofluorescence (mIF) demonstrated decreased phosphorylation of ERK and accumulation of p62 indicating a pharmacodynamic effect. Reverse phase protein array (RPPA) also confirmed MAPK pathway inhibition, with a decrease in pERK, and autophagy inhibition, with decreased autophagy signature score, within the tumor compartment of on-treatment biopsies compared to baseline. Single nucleus RNAseq with protein expression activity extrapolated using ARACNe (Algorithm for the Reconstruction of Accurate Cellular Networks) and VIPER (Virtual Inference of Protein-activity by Enriched Regulon) pipelines identified five distinct tumor subpopulations. Comparing prevalence of tumor clusters before and after treatment identified differential responses to treatment. Gene set enrichment analysis revealed that the most sensitive tumor cluster, tumor cluster 2, was enriched in autophagy and MAPK signaling. Pre- treatment frequency of tumor cluster 2 was higher in patients who experienced a clinical benefit and independently correlated with PFS. Among the non-tumor cell clusters identified, patients with clinical benefit also had increased immune cell infiltration. To explain the discordance in efficacy with this combination in preclinical models and humans, we interrogated the preclinical tumor model systems for tumor heterogeneity. Single cell RNA-Seq analysis of tumors from Kras LSL.G12D/+ ; p53 LSL.R172H/+ ; Pdx1-Cre tg/+ (KPC) mice treated with the combination confirmed presence of similar treatment-sensitive and treatment-resistant clusters, with the former being enriched for MAPK and autophagy signatures. Parallel analysis of cell lines previously shown to be sensitive to combination MEK and autophagy inhibition, including MiaPACA2 and PANC1, showed they were mostly enriched in the treatment-sensitive human cluster 2 phenotype but not treatment-resistant clusters. Our results suggest that some preclinical models may overrepresent subtypes which are sensitive to MEK and autophagy inhibitors and thus do not recapitulate the extent of intertumoral heterogeneity in human PDAC. Observed preclinical responses likely represented elimination of treatment-sensitive cluster(s), with the remaining clusters responsible for maintained resistance. Citation Format: Brian W Labadie, Aleksandar Z Obradovic, Sarah Sta Ana, Michael May, Naomi Sender, Isabelle V Ross, Monica Rodriguez-Mendez, Urszula Wasko, Lorenzo Tomassoni, Li Li, Manan Vij, Sinan Abuzaid, Julia Wulfkuhle, Winston Wong, Ilenia Pellicciotta, Benjamin Izar, Emanuel F Petricoin, Aik Choon Tan, Kenneth P Olive, Alex G Raufi, Gulam A Manji. PDAC tumors from patients treated with combination MEK and autophagy inhibition reveal a treatment sensitive epithelial subtype that is enriched in murine models [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2) nr C023.
oncology