Expression of Genes Associated with Antifolate (Pemetrexed) Activity by LUAD Subtype.

Development of the 48-gene LUAD nearest centroid classifier

Relationship of the expression of individual genes (ATIC, GART, DHFR, MTHFD1L or TYMS) with survival.

48 genes included in the new LUAD nearest centroid classifier and associated AF-PRS signature.

RNA Sequencing Quality Control Criteria

Expression Matrix of Patient Tumor Sample RNAseq

Abstract Introduction: MSI-H and mismatch repair deficient (dMMR) tumors are associated with favorable immune checkpoint inhibitor (ICI) responses (PMID: 3326454). However, up to 95% of CRC tumors are MSS/proficient MMR (pMMR) leading to poorer prognosis and treatment outcomes than MSI-H/dMMR patients (PMID: 12867608, 26028255). In MSS/pMMR metastatic CRC, multiple combination therapies are being investigated despite a lack of biomarkers to guide therapy. Therefore, we developed an RNA-based MSS-PRS with a primary objective to identify MSS tumors that have MSI-H/dMMR molecular features indicative of ICI response. Experimental Procedures: The MSS-PRS was developed using the TCGA CRC cohort (COAD; n=268). Training labels were assigned using the MSI Mantis score and mutation status from 8 genes with reported MSI association. Samples called “activated” had both a gene mutation and an MSI Mantis score > 0.4 (i.e., MSI-H), and samples called “non-activated” were wild type for all genes and had MSI Mantis score < 0.4. All other samples were designated “ambiguous” and excluded from training. Two thirds of the non-ambiguous samples were assigned to the classifier training set and all remaining samples were assigned to the test set. Using ClaNC software (PMID: 16269418) and cross-validation in the training set, a nearest centroid classifier was developed from a set of high mean, high variance candidate genes to select an optimal gene set to separate activated and non-activated groups. The classifier performance was evaluated in the test set. Results: The MSS-PRS contained 112-genes enriched for mismatched DNA binding, DNA damage repair, PD-L1, innate and adaptive immune response, cellular immunity, and cytokines. In the test set, the classifier called 44 of 46 MSI-H samples activated; of the remaining 77 samples that were MSS tumors, 38 were called activated and 39 were called not activated. Visual inspection of heat maps of tumor by MSI/MSS-PRS status and their association with TMB and CRC-related mutations suggest greater differences between MSS tumors called activated and MSI-H tumors than between MSS tumors called activated and MSS tumors called not activated. In contrast, immune marker expression profiles in MSS tumors called activated were markedly more similar to MSI-H tumors compared to MSS tumors called non-activated. Summary and Conclusions: Herein we described the development of a novel MSS-PRS that captures an MSI-H/dMMR-like molecular phenotype in a subset of patients with MSS tumors despite their lack of high TMB or overt MSI defects. Based upon these initial findings, further development of the MSS-PRS and its clinical validation as a tool to select patients with MSS tumors who may benefit from ICI-containing treatment regimens is warranted. Citation Format: Kirk Pappan, Gregory Mayhew, Jonathan Shepherd, Yuelong Guo, Kirk Beebe, Joel Eisner, Michael Milburn. Development of a novel RNA-based microsatellite stable predictive response signature (MSS-PRS) to identify MSS colorectal cancer (CRC) patients with a microsatellite instability-high (MSI-H) molecular phenotype [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1040.

Table names for supplementary tables

548 Background: Interest in FGFR-targeted (FGFRi) therapies for UC or pan-tumor use is growing (ongoing clinical studies include erdafitinib (NCT05316155; NCT04172675; NCT03390504; NCT04083976), LOXO-435 (NCT05614739), and pemigatinib (NCT03914794) following accelerated approval of erdafitinib in locally advanced/metastatic (m) post-chemotherapy UC patients with FGFR2/3 (i.e., DNA mutations and fusions) alterations (ALT). However, ALT status may not fully capture all who may show clinical benefit as responses are observed in both ALT (+) and (-) patients. Improved test strategies are needed to identify patients most likely to respond to FGFRi. FGFR-PRS is an RNA-based signature developed to capture the biology of FGFR-active tumors independent ALT status and is intended to be used as a diagnostic (Dx) to identify a broader patient population likely to respond to FGFRi. Methods: Known oncogenic FGFR3 ALTs (S249C, R248C, FGFR3c-Y373C, FGFR3c-G370C, FGFR3:TACC3 and FGFR3:BAIAP2L1) were used as training labels for nearest centroid classifier development as described in Dabney et al, 2005 ( https://doi.org/10.1093/bioinformatics/bti681 ) using 2/3 of data from the TCGA BLCA dataset, predominantly comprised of muscle-invasive bladder cancer patients and a few mUC patients, for training. The resulting 80-gene signature was applied to the remaining 1/3 of TCGA and a separate BACI mUC dataset (Rose et al, 2021; https://doi.org/10.1038/s41416-021-01488-6) as test sets. In vitro drug sensitivity was evaluated using data from the Genomics of Drug Sensitivity in Cancer database (GDSC; https://www.cancerrxgene.org/ ). Differential gene expression (DGE) results were analyzed using WebGestalt 2019 (https://www.webgestalt.org/). Results: Thirteen and 18% of the TCGA and BACI UC cohorts were ALT (+), whereas 49% and 48% were FGFR-PRS (+) independent of ALT status. About half of the patients in either cohort were ALT (-)/ FGFR-PRS (-) and only 1% were ALT (+)/FGFR-PRS (-). FGFR 1-3 selective drugs had strong inverse correlations between FGFR-PRS score and IC50 across 18 GDSC bladder cancer cell lines. FGFR-PRS (+) samples had enhanced co-expression of FGFR3, FOXA1, and SHH and selected multiple differentiation ontologies. Conclusions: FGFR-PRS (+) captured most ALT (+) tumors and an additional 2X more with similar FGFR pathway activation. FGFR-PRS (+) tumors were associated with gene enrichments for ontologies linked to FGFR3 signaling. The correlation of FGFR-PRS score with in vitro FGFRi activity provided initial utility of the signature, which is undergoing clinical evaluation in the ALAMANCE retrospective study of UC patients treated with FGFRi or other standard-of-care therapies. Analytical validation is ongoing to support FGFR-PRS for use as a clinical trial assay and eventual Dx.

AbstractPurpose: The Genomic Analysis of High-Risk Non–Muscle-Invasive Bladder Cancer (GARNER) study investigated FGFR alteration (ALT) frequency and the clinical outcome relationship with Bacillus Calmette–Guérin (BCG) treatment in high-risk non–muscle-invasive bladder cancer (HR-NMIBC). An FGFR predictive response signature (FGFR-PRS) was discovered that identifies patients with an activated FGFR pathway who could potentially benefit from FGFR-targeted therapy beyond those who are FGFR ALT (+). Experimental Design: Pretreatment tumor samples and clinical data were analyzed from 582 BCG-treated patients with HR-NMIBC. FGFR-PRS was discovered using a separate bladder cancer dataset and applied to the GARNER and other bladder cancer cohorts. FGFR-PRS was also applied to in vitro data from urothelial cancer cell lines treated with FGFR-active agents. Results: A total of 31% of pretreatment GARNER HR-NMIBC tumors were FGFR ALT (+), but this was not significantly associated with BCG response. For the subset of patients with paired pre- and post-BCG treatment samples, nearly one-third of pretreatment ALT (+) patients were ALT (−) posttreatment. FGFR-PRS identified patients with an activated FGFR pathway and identified approximately twofold additional patients compared with ALT status alone, and this increase was similar across tumor stage. A positive relationship between tumor growth inhibition and FGFR-PRS score was shown in bladder cancer in vitro models treated with FGFR-active agents. Conclusions: These data provide support for FGFR-targeted therapy use in FGFR ALT (+) HR-NMIBC and describe tumors with shared FGFR pathway–activated biology that is FGFR ALT (−) but FGFR-PRS (+). The latter suggests a broader potential patient population for FGFR-targeted therapy, which will require subsequent validation in patients treated with FGFR-targeted therapy.

549 Background: Immune checkpoint inhibition (ICI) shows great promise, including approval of pembrolizumab, nivolumab and avelumab in urothelial cancer (UC). However, the overall percentage of UC patients with clinical benefit is relatively small, despite biomarker use to select patients (e.g., PD-1 expression and TMB). Recently, several accelerated approvals (atezolizumab and durvalumab) were subsequently withdrawn following failure to show improved survival compared to standard of care (SOC) treatment in Phase 3 +/- biomarker use. To address the need for better ICI biomarkers, we developed an RNA-based predictive response signature (ICI-PRS) with the primary objective of being prognostic for ICI-treated patients, but not for SOC treatment. Methods: A survival-based model trained on 2/3 of the combined IMVigor210 dataset (n=298 locally advanced or metastatic UC patients treated with atezolizumab) and TCGA BLCA cohort (n=405 patients treated with non-ICI SOC) identified genes associated with survival across these data sets (prognostic agnostic to therapy) and genes distinctly associated with survival with ICI treatment (candidate ICI-predictive). Candidate ICI-predictive genes were developed into a model that was applied to the remaining IMVigor210 and TCGA BLCA samples, UC-GENOME urothelial carcinoma samples, and an ICI-treated melanoma dataset (Gide et al, 2019 : DOI: 10.1016/j.ccell.2019.01.003). Associations with tumor mutation burden (TMB) were examined. Expanded models incorporating this signature with additional immune markers were also investigated. Results: The ICI-PRS shows consistent associations with overall survival (OS) dependent on ICI treatment status. A higher hazard ratio (HR) was found for cohorts with ICI-therapy [HR (95% CI): IMVigor test 1.23 (0.96-1.57); UCGENOME 1.35 (0.99-1.85); Gide 1.50 (1.02-2.2)]. By comparison, no association was seen in the TCGA test set in which patients received non-ICI SOC treatment (HR = 0.96 (0.73-1.3). Patients with high TMB had extended survival in both IMVigor210 and non-ICI SOC TCGA datasets (P = 0.02 and 0.001). Conclusions: Herein we describe the development of a novel ICI-PRS and initial demonstration that it provides ICI-treatment specific prognostic value for bladder cancer patients. ICI-PRS status was associated with extended survival in patients treated with anti-PD-L1 but not non-ICI SOC therapy. These results support the further evaluation of the ICI-PRS and potential development as a diagnostic test for selecting UC patients most likely to benefit from ICI therapy.

140 Background: PRL-02 is a novel long-acting intramuscular (IM) depot injection prodrug of abiraterone (AA) that was designed to provide better safety through lower and less variable plasma exposures. Clinical results show that PRL-02 has a potentially superior therapeutic index compared to oral AA due to preferential inhibition of CYP17 lyase and lack of induction of progesterone, a known oncogenic driver, and tissue and lymphatic distribution. PRL-02 markedly depressed testosterone (T) through 14 wks in a castrate monkey model treated with a single dose. We present an interim report of data from an ongoing phase 1 study that includes the provisional RP2D and data from patients (pts) with prior exposure to enzalutamide (enza) recruited in either dose escalation or a dedicated expansion cohort. Methods: The study has a standard 3+3 dose escalation design (dose limiting toxicity [DLT] period = 28 days) with IM PRL-02 depot administered every 84 days. Pts with biochemical relapse, mCSPC, or mCRPC and ongoing GnRH analogue therapy (in absence of orchiectomy) with screening T < 50 are included. Pts with prior treatment with a CYP17 inhibitor and concurrent treatment with an AR blocking agent are excluded. Two expansion cohorts are recruiting: prior exposure to AA or enza. Results: Dose escalation: 22 pts (9 mCRPC, 12 mCSPC, 1 CSPC), median age 68, were treated at 5 dose levels: 180/360/720/1260/1800 mg (n=3/3/4/6/6) with median time on treatment (TOT) in months (mo) of 2.17 (1.88-2.20), 2.93 (2.37-23.52), 19.97 (1.91-21.48), 16.18 (11.81-19.87), and 10.66 (4.74-17.11), respectively. At 720 and 1260 mg, the median baseline T was 6.7 ng/dL and T suppression without significant changes in cortisol due to inhibition of CYP17 hydroxylase was observed in all pts. Minor transient progesterone elevations were observed at 1800 mg only. 3 of 4 pts at 720 and all pts at 1260 mg achieved a prostate specific antigen (PSA) response and had either a 90% reduction in T or values ≤ 1. One pt (1800 mg, CRPC heavily pretreated) progressed on treatment. No DLTs, treatment related serious adverse events (SAEs), or AEs related to PRL-02 with severity greater than CTCAE G2 were observed. In the prior enza expansion cohort 8 pts have been treated. Of the 10 pts with prior enza across both the escalation and expansion cohort, 8 had no prior chemo. Median TOT is currently 4.34+ mos. 3 pts progressed. All 7 active pts demonstrated either a PSA decrease from baseline or a flattening of PSA trajectory, including 2 with early PSA50s. Conclusions: PRL-02 was well tolerated. The RP2D of 1260 mg was selected over 1800 mg due to the absence of progesterone rises at 1260 mg. PRL-02 produces durable dose dependent T suppression and associated PSA response. The phase 1 data encourages continued investigation of PRL-02 for pts who have developed androgen receptor pathway inhibitor (ARPI) resistance. Clinical trial information: NCT04729114 .

High-dose (HD) IL-2 was the first immuno-oncology agent approved for treating advanced renal cell carcinoma and metastatic melanoma, but its use was limited because of substantial toxicities. Multiple next-generation IL-2 agents are being developed to improve tolerability. However, a knowledge gap still exists for the genomic markers that define the target pharmacology for HD IL-2 itself. In this retrospective observational study, we collected PBMC samples from 23 patients with metastatic renal cell carcinoma who were treated with HD IL-2 between 2009 and 2015. We previously reported the results of flow cytometry analyses. In this study, we report the results of our RNA-sequencing immunogenomic survey, which was performed on bulk PBMC samples from immediately before (day 1), during (day 3), and after treatment (day 5) in cycle 1 and/or cycle 2 of the first course of HD IL-2. As part of a detailed analysis of immunogenomic response to HD IL-2 treatment, we analyzed the changes in individual genes and immune gene signatures. By day 3, most lymphoid cell types had transiently decreased, whereas myeloid transcripts increased. Although most genes and/or signatures generally returned to pretreatment expression levels by day 5, certain ones representative of B cell, NK cell, and T cell proliferation and effector functions continued to increase, along with B cell (but not T cell) oligoclonal expansion. Regulatory T cells progressively expanded during and after treatment. They showed strong negative correlation with myeloid effector cells. This detailed RNA-sequencing immunogenomic survey of IL-2 pharmacology complements results of prior flow cytometry analyses. These data provide valuable pharmacological context for assessing PBMC gene expression data from patients dosed with IL-2- related compounds that are currently in development. The Journal of Immunology, 2024, 213: 29-39.- 39.

HCI-009 PDX experimental principal component analyses. (A-B) RNAseq principal component analyses (PC) for HCI-009 PDX experiments. (A) HCI-009 PDX tumor portions were injected into the mammary fat pads of female NSG mice supplemented with cellulose or DHT (dihydrotestosterone); n=5-7 mice per group. (B) HCI-009 PDX tumor portions were injected into the mammary fat pads of female NSG mice. Mice were administered vehicle or 150mg/kg/day Sevi (seviteronel); n=4-5 mice per group.

141 Background: PRL-02 (abiraterone decanoate) is a novel, long-acting IM depot prodrug of abiraterone. In non-clinical models, PRL-02 provided high and durable concentrations of prodrug and abiraterone to target tissues including adrenal glands, lymph nodes and bone. Single doses of PRL-02 suppressed testosterone (T) through 14 weeks in a castrate monkey model to concentrations comparable to oral abiraterone acetate (AA) with lower and less variable plasma abiraterone exposures. Clinically, PRL-02 blocks androgens with minimal increases in mineralocorticoids or depletion of glucocorticoids via inhibition of CYP17 lyase and minimal inhibition of CYP17 hydroxylase. PRL-02 has the potential for a superior therapeutic index and safety profile compared to oral AA. Methods: Phase 1 is a standard 3+3 design intended to identify a recommended phase 2 dose (RP2D). Pts with metastatic castrate resistant or sensitive prostate cancer (mCRPC/mCSPC) and a screening T of 2 - 50 ng/dL were administered IM PRL-02 every 12 weeks with daily oral dexamethasone. Results: As of 8Sep22, 17 pts (6 mCRPC, 11 mCSPC) were treated across 5 dose cohorts (180, 360, 720, 1260, 1800mg). Generally, there was a dose-proportional increase in abiraterone concentrations following a single dose of PRL-02 with a Tmax of 14 - 28 days and a plasma half-life of 18.3 days. The median baseline T level was 7.45ng/dL. Among pts dosed at 720mg and above, 9 of 11 had a 90% reduction in T or values ≤ 1ng/dL at day 28, including 2 pts with T≤LLOQ of 0.1 ng/dL, and PSA50 responses were observed in 8 of 10 with post-baseline results. There were no treatment related serious adverse events (AEs) or dose limiting toxicities. G3 AEs related to PRL-02 included hip and shoulder pain. G2 related AEs included fatigue, decreased appetite, insomnia and hot flush; symptoms of mineralocorticoid excess were not reported. Minimal and transient changes in ‘up-stream’ steroids (e.g., progesterone (P) and corticosterone (C)) were observed through the 1800mg dose. Although serial radiology was not prospectively required, there was radiographic improvement in 6 pts with data available. Conclusions: PRL-02 was well tolerated. Dose-dependent T suppression was associated with clinical benefit including PSA responses and radiographic improvement. Based on a historical comparison, the levels of P and C are significantly lower than seen with AA + prednisone which appear to be due to greater CYP17 lyase selectivity. The 1260mg or 1800mg dose will be the RP2D. Clinical trial information: NCT04729114 . [Table: see text]

Overlap between RNAseq datasets from MDA-MB-453 AR-High cells and HCI-009 PDX mice treated with Sevi

Supplementary Figure S1: Additional expression data and comparisons of AR expression by subtype, age, sex, and common molecular alterations. Supplementary Figure S2. Tumor growth plots and mice weights for in vivo studies. Supplementary Figure S3: Heat map of down-regulated genes from the GBM 26 RNA-seq experiment. Supplementary Figure S4: Effects of seviteronel on dsDNA break repair in AR-positive and negative models post-radiation. Supplementary Figure S5: Effects of enzalutamide on dsDNA break repair in AR-positive and negative models post-radiation. Supplementary Figure S6. Effects of enzalutamide and seviteronel on alkaline tail moment in AR-positive models post-radiation. Method: alkaline comet assay.

Gene expression change following Sevi treatment in HCI-009 PDXs

Sevi method of action. (A) Schematic of the seviteronel (Sevi) mechanism of action.

Genes correlated with AR expression in MDA-MB-453 cells