Search

Background: Colorectal liver oligometastases (CLO) represents a subset of metastatic colorectal cancer (mCRC) patients who may be eligible to undergo curative intent resection which can improve survival. However, given the heterogeneity of this population, prognostic stratification of CLO patients is poorly understood, with a highly variable 5 year survival rate of 25-58%. A three-gene prognostic biomarker based on pathogenic mutations in RAS (KRAS or NRAS), SMAD4, and TP53 was previously described. Here, we validate the prognostication of this three-gene biomarker in a CLO real world data (RWD) cohort. Methods: We leveraged clinico-molecular real world data to develop a retrospective, de-identified cohort of CLO patients for the assessment and development of prognostic biomarkers. Inclusion criteria include curative intent liver resections with negative or unknown surgical margins, tumor samples collected from the colon or liver, metastasis to only the liver at time of sample collection, and minimum 30 days follow up following resection. All samples were profiled with the Tempus xT assay (DNA-seq of 595-648 genes at 500x coverage). Real-world time to progression (rwTTP) was used to assess the prognostic value of the three-gene biomarker using a Cox proportional hazard model, treating gene mutations as both a continuous variable (the total number of genes mutated 0-3) and as categorical variables, while also controlling for neoadjuvant therapy, adjuvant therapy, and metachronous vs. synchronous colorectal cancer. Results: We identified 276 patients that met the inclusion criteria for this study. Of these patients, 27 had three genes with mutations, 85 with two, 155 with one, and 9 with zero mutations. The three-gene biomarker represented as a continuous variable had a rwTTP hazard ratio (HR) of 1.42 (95% confidence interval 1.15-1.77, p=0.001). Comparing three mutations to zero mutations resulted in a HR of 10.60 (95% CI 2.2-50.8, p=0.003, n=36), HR=2.55 (95% CI 1.54-4.22, p=0.0003, n=182) compared to one mutation, and HR=2.47 (95% CI 1.44-4.22, p=0.001, n=112) compared to two mutations. Median rwTTP for patients with mutations in all three genes was 3.7 months compared to 31.0 months for patients with zero mutations from the three-gene biomarker, and 10.0 months for all patients in the cohort. Conclusions: A novel clinico-molecular RWD cohort allowed independent rwTTP validation of a three-gene biomarker in the CLO setting. Although future prospective studies are warranted, the significance of the three-gene biomarker as a prognostic factor suggests that mutation status of RAS, SMAD4, and TP53 may be helpful to guide treatment decisions surrounding curative surgery. Citation Format: Alia Zander, Caroline Epstein, Kabir Manghnani, Ben Terdich, Sun Hae Hong, Justin Guinney, Halla Nimeiri, Martin Stumpe, Timothy Taxter, Kyle A. Beauchamp. Clinico-molecular real world data demonstrates prognostic significance of a three-gene biomarker for colorectal liver oligometastases [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr A003.

2018 Background: In 2021, the WHO revised its classification of central nervous system tumors (WHO CNS5) around IDH status and inclusion of key somatic alterations in addition to histopathological traits. While providing a more specific classification system for patients, the guidelines introduce new logistical challenges for pathologists, relying on multi-modal data for accurate classification. In this study, we use a combined clinical/molecular real world dataset to reclassify a cohort of adult diffuse gliomas and evaluate prognostic impact using real-world overall survival (rwOS). Methods: We retrospectively analyzed a de-identified dataset of 2,703 adult diffuse glioma samples profiled with the Tempus xT assay (DNA-seq of 595-648 genes at 500x coverage, whole-exome capture RNA-seq). Original diagnoses were identified from sample pathology reports. We assessed mutation status for genes relevant to WHO CNS5 classification ( IDH1/2, ATRX, and TERT) and copy number alterations for genes CDKN2A/B and EGFR, as well as arms 1p, 19q and chromosomes 7 and 10. Samples were excluded if the original diagnosis or molecular findings indicated a diffuse midline or pediatric glioma. Necrosis and microvascular proliferation were inferred from gene expression profiles using machine learning. rwOS was defined as time from original diagnosis until death. To account for left truncation, samples were only considered at risk of death after study entry ( e.g., date of sequencing if sequenced as part of clinical care). Results: Using relevant clinicopathological traits and genomic alterations, we assigned WHO CNS5 labels to all samples—512 astrocytoma, IDH -mutant; 186 oligodendroglioma, IDH -mutant, and 1p/19q co-deleted; and 2,005 glioblastoma, IDH -wildtype (IDH- wt). We further stratified astrocytoma, IDH -mutant samples by grade, resulting in 308 classified as grade 4. Of the samples with an original diagnosis of astrocytoma, oligodendroglioma, or glioblastoma, 13.9% changed under the WHO CNS5 guidelines, including 166 glioblastomas re-classified as grade 4 astrocytoma, IDH -mutant, and 125 astrocytomas re-classified as glioblastoma, IDH -wt. WHO CNS5 reclassification resulted in more accurate prognostic stratification (rwOS) than the original diagnosis (likelihood ratio test; P < 8e-32). For example, the observed hazard ratio (HR) of an original astrocytoma diagnosis (versus glioblastoma) (Cox PH Model; HR = 0.42[CI:0.32,0.56]) was less extreme than the observed HR of an astrocytoma, IDH -mutant diagnosis (versus glioblastoma, IDH -wt) under WHO CNS5 (Cox PH Model; HR =0.08[CI:0.04,0.19] grades 2-3; HR = 0.18[CI:0.12,0.26] grade 4). Conclusions: This work highlights the utility of comprehensive molecular profiling in classifying patients with adult diffuse gliomas according to the newest WHO CNS5 guidelines, and confirms the improved prognostic stratification within a retrospective real-world dataset.

3077 Background: While targeted DNA-seq can detect clinically actionable fusions in tumor tissue samples, technical and analytical challenges may give rise to false negatives. RNA-based, whole-exome sequencing provides a complementary method for fusion detection, and may improve the identification of actionable variants. In this study, we quantify this benefit using a large, real-world clinical dataset to assess actionable fusions detected from RNA in conjunction with DNA profiling. Methods: Using the Tempus Research Database, we retrospectively analyzed a de-identified dataset of ̃80K samples (77.4K patients) profiled with the Tempus xT assay (both DNA-seq with fusion detection in 21 genes and whole exome capture RNA-seq). Only patients that had successful RNA- and DNA-seq were included. Fusions were detected using the Tempus bioinformatic and clinical workflow. Candidate fusions were filtered based on read support thresholds, fusion annotation ( i.e., breakpoints, reading frame, conserved domains), and manual review. OncoKB was used to select fusion alterations in levels 1 and 2 and to identify those indication-matched to targeted therapies. Results: We identified 2118 level 1 and 2 fusion events across 1945 patients across 20 different cancer types. Most fusions were observed in non-small cell lung cancer (NSCLC) (25%) and biliary cancer (9%) samples. Of the 2118 fusion events, 29.1% (616) were detected only through RNA-seq while 4.8% (101) of the events were identifiable only through DNA-seq. Notably, 69.4% of fusions in low-grade glioma and 58.2% in sarcomas were detected only by RNA-seq. When evaluating specific gene fusion events, RNA-seq consistently improved the detection of fusions compared to DNA-seq alone (Table) across all cancer types. A total of 1106 fusions were classified as targetable by OncoKB indication-matched therapies with 19% (214) of these identifiable through RNA-seq alone, 5% (54) by DNA-seq alone, and 76% (838) identifiable through RNA- and DNA-seq. Overall, fusions identified through RNA-seq alone led to a 24% increase in the number of patients who were eligible to receive matched therapies (214 / 892). This included imatinib for patients with CML/BLCL (69.8%), crizotinib for NSCLC (40.3%) and entrectinib for NTRK and ROS1 fusions (32.5%). Conclusions: The addition of RNA-seq to DNA-seq significantly increased the detection of fusion events and ability to match patients to targeted therapies. Results support consideration of combined RNA-DNA-seq for standard-of-care fusion calling. [Table: see text]