Search

8577 Background: Predicting variable therapeutic responses that are driven by the genetic and transcriptomic heterogeneity of Small Cell Lung Cancer (SCLC) offers an opportunity for implementing precision therapies within the cancer cells and tumor microenvironment (TME). MERU was a Phase III study of Rovalpituzumab Tesirine (Rova-T) as maintenance therapy following first-line platinum-based chemotherapy in participants with extensive stage SCLC. In this study, we comprehensively analyzed the baseline genomic data in the MERU cohort to interrogate SCLC’s heterogeneous TME and tumor-intrinsic molecular and genetic drivers for therapeutic vulnerabilities. Methods: RNA-seq and Whole-Exome-Sequencing (WES) data were collected from archival tumor samples of 306 of 740 subjects enrolled in MERU. RNA-seq reads were aligned with STAR and quantified for gene expression by HTSeq. WES reads were analyzed for somatic mutation and copy number variation using AbbVie’s tumor-only WES pipeline. TME heterogeneity was evaluated using gene-set variation analysis of pan-cancer TME gene signatures. SCLC subtyping was performed using expression of 4 transcriptional factors (TF): ASCL1, NEUROD1, POU2F3, and YAP1. A computational framework of mapping MERU transcriptome expression profile to Cancer Dependency Map (DepMap) dataset was developed to synthetically screen MERU samples’ drug sensitivity. Results: TF subtyping reveals high prevalence of SCLC-A and -N subtypes in the MERU cohort, consistent with high expression of DLL3 in these two neuroendocrine subtypes. Correlation of TME gene signature scores identified two distinct clusters in the MERU cohort with correspondingly polarized immune-suppressive and -inflamed phenotypes. The pro-inflammatory score combining IFN-gamma and TGF-beta signatures, predicted prognosis in the MERU cohort better than the previously reported SCLC-I signature (Hazard Ratio: 0.71 [95% CI 0.38-1.3] vs. 1.29 [95% CI 0.65-2.6]). WES analysis identified high prevalence of TP53 and RB1 mutations, in line with the reported prevalence of these genetic drivers in SCLC. The clinical characteristics including gender, smoking status, and prior treatment, are not significantly associated with either TF or TME subgroup. The computational drug screen framework maps 83% of MERU samples to DepMap SCLC cell lines. The correlation of subtype TF expression with drug sensitivity was highly concordant between MERU and DepMap. Collectively, this approach demonstrated that molecular subtyping can be leveraged to broadly predict drug response in SCLC patients. Conclusions: Our comprehensive genomic analysis of the MERU cohort provides new insights into SCLC heterogeneity from both tumor-intrinsic and tumor-immune interaction perspectives and shall contribute to the development of predictive biomarkers and therapeutic opportunities for SCLC.

Disclosures: All authors are employees of AbbVie. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication. Ovarian cancer describes a group of malignancies that remain a serious threat to women's health and claim more than 14,000 deaths in the US each year. Among these, high-grade serous ovarian cancers (HGSC) represent the most common and aggressive form, and tumor recurrence is near-universal following initial response to carboplatin and paclitaxel. While multiple treatment options are available for platinum-sensitive recurrent HGSC, such as continued use of carboplatin/paclitaxel with or without bevacizumab, as well as several FDA-approved PARP inhibitors, there are few effective treatment options for platinum-resistant HGSC. Its inherent heterogeneity, characterized by genomic instability and numerous DNA copy number aberrations, poses challenges to the discovery of novel therapeutic targets. Patient-derived xenograft (PDX) models are an important platform for target identification and efficacy testing of therapeutic agents, as they sustain the intratumoral heterogeneity observed in advanced human tumors more faithfully than established cell lines. The isolation of distinct tumor cell populations by flow cytometry and subsequent testing of their tumorigenic potential can identify tumor-initiating cells (TICs), which have the potential to propagate the tumor over multiple generations in immune-deficient NOD/SCID mice. Thorough characterization of these TIC populations by transcriptome profiling and flow cytometry has identified TIC-enriched cell surface proteins that are targetable by antibody drug conjugates (ADCs), including EFNA4 in mixed Müllerian and ovarian cancers and PTK7 in triple-negative breast and ovarian cancers. ADCs consist of a monoclonal antibody directed against a cell surface epitope linked to a cytotoxic agent, such as auristatin, maytansinoid, or pyrrolobenzodiazepine (PBD). Here, we describe the generation and characterization of a PDX bank for ovarian cancers that led to the identification of dipeptidase 3 (DPEP3) as a TIC-associated target in HGSC. We show that DPEP3 is enriched in the TIC fraction of platinum-sensitive and platinum-resistant HGSC PDX models, where it localizes to the plasma membrane and is detected by flow cytometry and immunohistochemistry (IHC). In contrast, DPEP3 expression is low or absent in most normal adult tissues, thus providing a therapeutic window for an antibody-based therapeutic agent. In order to target DPEP3-expressing ovarian cancer cells, we developed SC-003, an ADC consisting of a humanized monoclonal antibody linked to a PBD dimer via a cleavable linker. We show that SC-003 specifically binds to DPEP3-expressing cells and, upon internalization, elicits cytotoxicity via release of its PBD warhead following lysosomal degradation of the antibody component. A single dose of SC-003 induced tumor regression in DPEP3-positive HGSC PDX models, including platinum-resistant PDX models. Mechanistically, we show that the anti-tumor effect of SC-003 is mediated by a significant reduction in TIC frequency. Moreover, combination with an anti-PD1 antibody potentiated SC-003 efficacy in a syngeneic mouse model engineered to overexpress human DPEP3. In summary, these findings support the clinical development of SC-003 as a novel therapeutic agent for HGSC. Citation Format: Wolf R. Wiedemeyer, Sheila Bheddah, Christine Saechao, Dorothy French, Erik Huntzicker, Aaron Kempema, Julia Gavrilyuk, David Liu, Vikram Sisodiya, Alina He, Zhaomei Zhang, Monette A. Aujay, Kristyn Hayashi, Sandro Vivona, Xi Zhao, Kimberly Walter, Laura R. Saunders, Johannes Hampl, Shravanthi Madhavan, Marybeth Pysz, Alexander J. Bankovich, Holger Karsunky, Scott J. Dylla. SC-003, AN ANTIBODY-DRUG CONJUGATE TARGETING DIPEPTIDASE 3, EXHIBITS POTENT ANTI-TUMOR ACTIVITY IN PATIENT-DERIVED XENOGRAFT MODELS OF HIGH GRADE SEROUS OVARIAN CANCER [abstract]. In: Proceedings of the 12th Biennial Ovarian Cancer Research Symposium; Sep 13-15, 2018; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2019;25(22 Suppl):Abstract nr NT-113.