Pan-tumor survey of ROS1 fusions detected by next-generation RNA and whole transcriptome sequencing

BACKGROUND

Two ROS1 tyrosine kinase inhibitors have been approved for ROS1 fusion positive (ROS1+) non-small cell lung cancer (NSCLC) tumors. We performed a pan-tumor analysis of the incidence of ROS1 fusions to assess if more ROS1+ patients who could benefit from ROS1 TKIs could be identified.

METHODS

A retrospective analysis of ROS1 positive solid malignancies identified by targeted RNA sequencing and whole transcriptome sequencing of clinical tumor samples performed at Caris Life Science (Phoenix, AZ).

RESULTS

A total of 259 ROS1+ solid malignancies were identified from approximately 175,350 tumors that underwent next-generation sequencing (12% from targeted RNA sequencing [Archer]; 88% from whole transcriptome sequencing). ROS1+ NSCLC constituted 78.8% of the ROS1+ solid malignancies, follow by glioblastoma (GBM) (6.9%), and breast cancer (2.7%). The frequency of ROS1 fusion was approximately 0.47% among NSCLC, 0.29% for GBM, 0.04% of breast cancer. The mean tumor mutation burden for all ROS1+ tumors was 4.8 mutations/megabase. The distribution of PD-L1 (22C3) expression among all ROS1+ malignancies were 0% (18.6%), 1%-49% (29.4%), and ≥ 50% (60.3%) [for NSCLC: 0% (17.8%); 1-49% (27.7%); ≥ 50% (53.9%). The most common genetic co-alterations of ROS1+ NSCLC were TP53 (29.1%), SETD2 (7.3%), ARIAD1A (6.3%), and U2AF1 (5.6%).

CONCLUSIONS

ROS1+ NSCLC tumors constituted the majority of ROS1+ solid malignancies with four major fusion partners. Given that > 20% of ROS1+ solid tumors may benefit from ROS1 TKIs treatment, comprehensive genomic profiling should be performed on all solid tumors.

Pan-tumor survey of RET fusions as detected by next-generation RNA sequencing identified RET fusion positive colorectal carcinoma as a unique molecular subset

BACKGROUND

RET fusions are driver alterations in cancer and are most commonly found in non-small cell lung cancer and well-differentiated thyroid cancer. However, RET fusion have been reported in other solid tumors.

MATERIAL AND METHODS

A retrospective analysis of RET+ solid malignancies identified by targeted RNA sequencing and whole transcriptome sequencing of clinical tumor samples performed at Caris Life Science (Phoenix, AZ).

RESULTS

As of March 22, 2022, a total of 378 RET+ solid malignancies were identified in 15 different tumor types and carcinoma of unknown primary (CUP) that underwent next-generation RNA sequencing. RET+ NSCLC and RET+ thyroid cancer constituted 66.9% and 11.1% of the RET+ solid malignancies, respectively. RET+ colorectal adenocarcinoma and RET+ breast adenocarcinoma constituted 10.1% and 2.6%, respectively. The estimated frequency of RET fusions within specific tumor types were NSCLC 0.7%, thyroid cancer 3.1%, colorectal cancer 0.2% and breast cancer 0.1%. KIF5B (46.8%) was the most common fusion partner followed by CCDC6 (28.3%) and NCOA4 (13.8%) in RET+ solid tumors. KIF5B-RET was the dominant fusion variant in RET+ NSCLC, NCOA4-RET was the dominant variant in RET+ colorectal carcinoma, and CCDC6-RET was the dominant variant in thyroid cancer. The most common single gene alterations in RET+ tumors were TP53 (34.8%), RASA1 (14.3%) and ARIAD1A (11.6%). RET+ CRC had a high median TMB of 20.0 and were commonly MSI-H.

CONCLUSIONS

RET fusions were identified in multiple tumor types. With a higher median TMB and commonly MSI-H, RET fusion positive CRC may be a unique molecular subset of CRC.

Genomic profiling of rare undifferentiated sarcomatoid subtypes of pancreatic carcinomas for potential response to immunotherapy

741

Background: While pancreatic adenocarcinoma (PDAC) remains a leading cause of cancer-related deaths, the highly aggressive PDAC subtype of undifferentiated sarcomatoid carcinoma (USC) remains poorly characterized as it comprises only 2-3% of all PDAC histology. Previous case reports suggest that immune checkpoint inhibitors could be a promising treatment strategy for USC, but the prevalence of established predictive biomarkers of response are largely unknown in this unique subpopulation. We identified PDAC USC patient samples from a large dataset and performed comprehensive genomic profiling to determine the prevalence of biomarkers associated with response to immunotherapy. Methods: PDAC USC patient samples (N=43) underwent central pathology review to confirm this diagnosis and were compared to non-USC PDAC patient samples (N=5562). Next-generation sequencing of DNA and RNA was performed at Caris Life Sciences (Phoenix, AZ). PD-L1 expression was tested by IHC (SP142; Positive (+): ≥ 2+, ≥%5). A combination of IHC, NGS, and fragment analysis was used to assess deficient mismatch repair/microsatellite instability high (dMMR/MSI). High tumor mutational burden (TMB-High) was defined as ≥10 mutations/MB. Immune cell fractions of the tumor microenvironment were estimated by RNA deconvolution analysis using quanTIseq. Chi-square tests with Bonferoni correction for multiple comparisons were used to determine significance. Results: Among PDAC USC samples, KRAS (86% USC vs 90% non-USC, p = 0.31, q = 1), TP53 (86% vs 77%, p = 0.16, q = 1), and CDKN2A (18% vs 23%, p = 0.45, q = 1) were the most commonly mutated genes with a similar prevalence compared to non-USC histologies; however, KRAS was amplified more frequently (7% USC vs <1% non-USC, q = 0.006). Furthermore, more USC tumors were PD-L1+ (63% vs 16%, q < 0.001), while few USC samples were dMMR/MSI (2% USC vs 1% non-USC, p = 0.45, q = 1) or TMB-High (2% vs 2%, p = 0.89, q = 1). All USC tumors were deplete of lymphocytes, but many were differentially enriched (>5%) with neutrophils (85% vs 57%, q = 0.03) or M2 macrophages (52% vs 28%, p = 0.006, q = 0.06). Conclusions: This work represents the largest molecular analysis of PDAC USC samples to date. Our analysis uncovered a different prevalence of amplified KRAS and PD-L1 expression in USC as compared to other PDAC histologies amidst an immune desert lacking lymphocytes in the USC tumor microenvironment. These findings provide evidence for further investigation into combination therapy of KRAS inhibitors with immune checkpoint inhibitors to target these immune-imbalanced microenvironment landscapes.

Analysis of HLA gene expression in patients with dMMR/MSI-H colorectal carcinoma resistant to immune checkpoint inhibitors

202

Background: Large studies have identified immune checkpoint inhibitors (ICI) as an effective therapy for deficient mismatch repair/microsatellite instability high (dMMR/MSI-H) colorectal cancer (CRC). However, a subset of dMMR/MSI-H CRC patients exist that do not benefit from ICI and show rapid cancer progression within the first 6 months of therapy. Genetic alterations of the host immune system, including loss of β2M and single copy loss of HLA molecules, can contribute to innate resistance to ICI. In this study, we sought to analyze the role of expression of HLA genes and β2M as determinants of innate resistance to ICI by analyzing an extensive clinico-genomic database. Methods: Next-generation sequencing (NGS) of DNA (592-gene or whole exome) and RNA (whole transcriptome) was performed on CRC patient samples (n = 24,394) submitted to a CLIA-certified laboratory (Caris Life Sciences, Phoenix, AZ). dMMR/MSI-H was assessed by IHC/NGS. PD-L1 expression was tested by IHC (SP142; positive (+): ≥2+, ≥%5). Real world overall survival and treatment data were obtained from insurance claims data. Time-To-Next-treatment (TTNT) was calculated from start of ICI monotherapy to the start of next therapy, or death. Kaplan-Meier estimates were used for comparison. A composite signature of MHC II gene expression was tested for molecular associations. Immune cell infiltration was estimated by RNA deconvolution using quanTIseq. Statistical significance was determined using Fisher’s Exact/Mann Whitney/X2 tests. Results: We identified 1549 patients with dMMR/MSI-H CRC; 242 patients of these had received pembrolizumab or nivolumab. Using TTNT as a proxy for progression on treatment, we divided the patients into two cohorts: >180 days TTNT and <180 days TTNT. Following manual curation of the cases by two oncologists and limiting the patients to those who had received ≥ 2 doses of either ICI and at least 180 days of recorded follow up, we generated cohorts of 77 patients (>180 days TTNT; good responders) and 34 patients (<180 days TTNT; poor responders). High expression of HLA genes (HLA-DRB1, HLA-B, HLA-DQB1, HLA-DPB1, HLA-DPA1- p<0.02; fold change 2.1-3) was found in the good responder group. No β2M alterations were detected in either subgroup. After dividing the entire cohort into quartiles, patients in the highest quartile of expression for either CD74, HLA-DQA1, HLA-DQB1, HLA-DPB1 or HLA-DRB1 had improved survival when compared to lower expressors (p< 0.05). High MHC II signature expression was associated with an increased rate of PD-L1+ (34 vs 8%, p<0.0001) and infiltration of M1 macrophages (9.1 vs 4.9%, p<0.0001). Conclusions: We identify elevated expression of HLA genes involved in formation of the MHC-II complex as a potential biomarker of improved response to immunotherapy that could, if further validated, optimize patient selection for ICI in CRC.

Molecular characterization and clinical outcomes of pancreatic neuroendocrine tumors (pNENs) harboring PAK4-NAMPT alterations

649

Background: The mTOR inhibitor, Everolimus (EVE), is FDA-approved for the treatment of advanced PNENs on the basis of delay of progression. The RADIANT-3 trial showed an increase in PFS from 4.6 to 11 months compared to placebo with an ORR of only 5%. Prior studies suggest that targeting the aberrant expression of mTOR regulators p21 activated kinase 4 (PAK4) and nicotinamide adenine dinucleotide biosynthesis enzyme nicotinamide phosphoribosyltransferase (NAMPT) in PNENs sensitizes these tumors to EVE. To further qualify these observations, we queried a large real-world dataset of PNENs, characterizing the molecular and immune landscapes, as well as the clinical outcomes associated with aberrant PAK4 and NAMPT expression. Methods: 294 cases of PNENs were analyzed using Next Generation Sequencing (NextSeq) and Whole Exome and Whole Transcriptome Sequencing (NovaSeq) at Caris Life Sciences (Phoenix, AZ). For our analyses, we stratified our study cohort into four groups based on the median expression of PAK4 and NAMPT: PAK4-low/NAMPT-low, PAK4-low/NAMPT-high, PAK4-high/NAMPT-low and PAK4-high/NAMPT-high. Significance was determined using chi-square, Fisher-Exact or Mann-Whitney U, and p-values were adjusted for multiple comparisons (q < 0.05). Results: High prevalence of mutations in PTEN (10.71% vs 1.18%; p < 0.05, q > 0.05), a tumor suppressor of the mTOR pathway and high expression of genes activated in response to mTOR activation such as SLC2A1 (3.07-fold), PFKP (3.32-fold), SCD (2.70-fold), MVK (2.92-fold) and G6PD (2.58-fold) were observed in PAK4-high/NAMPT-high compared to the PAK4-low/NAMPT-low tumors (all q < 0.05). A congruent enrichment of PI3K/AKT/mTOR and glycolysis pathways by single-sample gene set enrichment analysis was observed in these tumors (all q < 0.05). When querying the immune landscape, we observed enrichment in inflammatory response, IL6/JAK/STAT3, IL2/STAT5 signaling pathways and immune checkpoint genes such as PDCD1 (5.14-fold), CD274 (2.84-fold), PDCD1LG2 (2.44-fold), CD80 (3.00-fold), CD86 (2.82-fold), IDO1 (1.92-fold), LAG3 (3.09-fold), HAVCR2 (2.66-fold) and CTLA4 (4.49-fold) in the PAK4-high/NAMPT-high tumors (all q < 0.05). Immune cell infiltration estimates revealed an increase in Neutrophils, NK cells and Tregs in the PAK4-high/NAMPT-high tumors (p < 0.05, q > 0.05). Conclusions: Our study demonstrates that PAK4-high/NAMPT-high PNENs are associated with distinct molecular and immune profiles. While the dual blockade of PAK4 and NAMPT has been reported to enhance the efficacy of EVE in PNENs, whether such a blockade would enhance the efficacy of immunotherapeutics warrants further investigation.

Characterizing colorectal cancer (CRC) carriers of the recessive MUTYH founders (G396D/Y179C) and the low-penetrance APC founder APC-I1307K mutation

210

Background: The clinical significance of CRC risk imparted by prevalent recessive mutations MUTYH-G396D and Y179C (Caucasian carrier rate 1/50) and the low-penetrance APC-I1307K mutation (Ashkenazi Jewish carrier rate ~1/15) is debated. Evidence supporting differential mutational spectrum, immune biology, and clinical outcomes in tumors that arise in carriers could lend support to their relevance. We characterized CRC carriers of MUTYH-G396D/Y179C and APC-I1307K, as well as those with concurrent second-hit mutations, in comparison to MUTYH or APC wild-type (WT) tumors. Methods: Retrospective review of patient samples (n = 13,896) submitted to a CLIA-certified laboratory (Caris Life Sciences, Phoenix, AZ) for next-generation sequencing (NGS) of DNA (592-gene or whole exome) and RNA (whole transcriptome). Deficient mismatch repair/high microsatellite instability (dMMR/MSI-H) was assessed by IHC/NGS. PD-L1 expression was tested by IHC (SP142; positive (+): ≥2+, ≥%5). Immune cell infiltration was estimated by RNA deconvolution using quanTIseq (Finotello, 2019). Presumed monoallelic (mMut) and biallelic mutations (bMut) included those with or without, respectively, a mutation or variant allele frequency ≥60%. Statistical significance determined by Fisher’s-Exact/Mann Whitney/X2 tests. *indicates raw p-value that was not significant after correction for multiple comparisons. Real world overall survival was obtained from insurance claims data, with Kaplan-Meier estimates used for comparison. Results: MUTYH-G396D/Y179C (119, 26 bMut) and APC-I1307K (11 mMut, ) alterations were identified, with all 3 founders being more common among male patients and associated with slightly increased median age at time of biopsy. MUTYH bMut had higher rates of genomic vs mMut 6.0%, p=0.02*; WT 11.0%, p=0.05*), and increased dendritic cell (p=0.04) and CD4 T-cell (p=0.007) proportions. APC bMut tumors were less frequently dMMR/MSI-H (2.6% vs mMut 18.%, p=0.11*; WT 15.5%, p=0.03*), TMB-H (2.6% vs mMut 27.3%, p=0.03*; WT 17.3%, p=0.02*), and PD-L1+ (0% vs mMut 18.2%, p=0.05*; WT 9.4%, p=0.04*), with enrichment of the canonical CMS2 subtype (44% vs 9% mMut, p<0.01; WT 11%, p<0.0001). In pMMR/MSS CRC, the prognosis of MUTYH founder carriers was superior to MUTYH WT tumors (HR 0.44, CI 0.25-0.78, p=0.004), with a non-significant trend toward greater sensitivity (HR 0.50, CI 0.21-1.2, p=0.). The prognosis of APC I1307K carriers was superior to APC WT tumors (HR 0.51, CI 0.34-0.76, p<0.001). Conclusions: Carrier status of MUTYH-G396D/Y179C and APC-I3017K founders may positively impact prognosis in patients with pMMR/MSS tumors. Differences may derive from distinct molecular signatures in tumors from MUTYH and APC founder mutations and warrant further investigation of these biomarkers in CRC.

SETD2 gene expression and the molecular landscape of colorectal cancer (CRC)

184

Background: SETD2, a key methyltransferase modulating histone 3 gene transcription, has been shown to act as a tumor suppressor gene by reducing oxidative stress, colonic inflammation and tumorigenesis in animal models. SETD2 gene expression has been reported to be significantly downregulated in CRC and linked with poorer patient survival. Additionally, SETD2 plays an important role in DNA repair and loss of function mutations have been associated with increased tumor mutational burden (TMB), mismatch repair (MMR) deficiency, and benefit from immunotherapy. Hence, we aimed to characterize the molecular features associated with SETD2 gene expression in CRC. Methods: 15,425 CRC tumors tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing on DNA (592 genes or whole exome sequencing), RNA (whole transcriptome sequencing), and IHC were analyzed. SETD2-high and -low expression were defined as ≥ top and < bottom quartile of SETD2 transcripts per million (TPM), respectively. Cell infiltration in the tumor microenvironment (TME) was estimated by QuantiSEQ. Mann-Whitney U and X2/Fisher-Exact tests were applied where appropriate, with P-values adjusted for multiple comparisons ( q < .05). Gene expression profiles were analyzed for transcriptional signatures predictive of response to immunotherapy (T cell-inflamed) and MAPK pathway activation (MPAS). Real-world overall survival information was obtained from insurance claims data and Kaplan-Meier estimates were calculated for molecularly defined patients. Results: SETD2 expression was significantly increased in left-sided/rectal compared to right-sided tumors (median TPM [mTPM] 55.8 vs 51.1, q < .001). SETD2 mutations were associated with reduced SETD2 expression in pMMR/MSS CRC (mTPM 37.9 vs 54.3 in WT, q < .001), although not statistically significant in dMMR/MSI-H CRC (mTPM 43.6 vs 51.7 in WT, P = .17). Compared to SETD2-high tumors, SETD2-low was associated with increased rates of TMB-high (10.4% vs 8.2%, P = .009), dMMR/MSI-H (7.3% vs 5.6%, P = .02), and PD-L1 IHC levels (4.4% vs 2.5%, q < .002). Only minor differences in gene mutation and copy number rates were observed between SETD2-high and -low tumors , whereas SETD2-high TME was associated with increased immune cell infiltration, including neutrophils, B cells, NK cells, M2 macrophages, and dendritic cells ( q < .05). SETD2-high was also associated with increased T cell-inflamed and MPAS signatures (152% and 159% median increase, respectively, q < .001), regardless of tumor MMR status. Among dMMR/MSI-H CRC, SETD2-high tumors receiving immune checkpoint inhibitors had longer time-on-treatment (HR: 0.39, 95% CI: 0.2-0.76, P = .005) and a trend towards longer OS (HR: 0.29, 95% CI: 0.07-1.2, P = .069). Conclusions: Our data show distinct immune biomarkers and TME cell infiltration associated with SETD2 gene expression in CRC. These findings support a key role for SETD2 in modulating anti-tumor immunity and TME.

Prognostic indicators of KRAS G12X mutations in pancreatic cancer

735

Background: We have studied the role of KRAS mutations in relation to the prognosis in patients with advanced pancreatic ductal adenocarcinoma (PDAC). KRAS is a well-described oncogenic driver in PDAC, with mutations identified in over 90% of cases, typically involving codon 12. The three predominant missense variants include G12D, G12V and G12R. PDAC has the highest rate of G12R mutations compared to other malignancies, comprising 15-20% of KRAS-mutated tumors. This study presents a new finding in the progression of advanced PDAC utilizing a large clinical and genomic database to further characterize the clinical features of pathogenic KRAS variants in PDAC with a focus on G12R. Methods: PDAC samples were tested using whole transcriptome sequencing (WTS; Illumina NovaSeq) and NextGen DNA sequencing (NextSeq, 592 Genes and NovaSEQ, WES) at Caris Life Sciences (Phoenix, AZ). Transcriptomic signatures including MPAS (MAPK activation score), T-cell inflamed score and tumor micro environment (TME) characterization were calculated on WTS data. Significance was determined by X2 and Fisher-Exact and p adjusted for multiple comparisons (q) was < 0.05 (Benjamini-Hochberg). Real-world overall survival (rwOS) was obtained from insurance claims data and calculated from tissue collection to last contact; time-on-treatment (TOT) was calculated from start to finish of specific treatments; comparison was done by Kaplan-Meier test. Results: A total of 5,555 patients with PDAC harboring either KRAS G12D (n = 2,671), G12V (n = 1,871) G12R (n = 904) or G12C (n = 109) variants were identified. The patients with KRAS G12R mutant tumors had significantly longer median real-world overall survival (mRWOS) compared to G12D (452 vs 358 days, HR 0.82, CI 0.74 – 0.9, p < 0.0001). This difference persisted regardless of treatment with FOLFIRINOX or gemcitabine plus nab-paclitaxel. There was no difference in outcomesbetween patients with KRAS G12R, G12V or G12C. PD-L1 expression was significantly lower in G12R than in G12C or G12D (13% vs 27% vs 19%,) while the prevalence of TMB-H and dMMR was comparable across isoforms. Conclusions: Patients with KRAS G12R variants has improved rwOS compared to G12D irrespective of the chemotherapy regimen administered. Immune profiling suggested that the immune contexture in G12R-driven tumors are distinct from G12D as reflected by reduced PDL1 staining, decreased levels of multiple checkpoint receptors. We aim to further explore the molecular basis for these differences with a focus on PI3K and MAPK pathways. Based on this data, survivorship studies in patients with advanced PDAC should consider reporting KRAS mutational status.

Molecular landscape and clinical implication of CCNE1-amplified esophagogastric cancer

440

Background: Cyclin E1 ( CCNE1) amplification ( CCNE1Amp) drives chromosomal instability (CIN) and is associated with immune cell exclusion, liver metastasis, therapeutic resistance, and poor survival in gastric adenocarcinoma (GA). However, the clinical relevance of CCNE1Amp in a larger real-world population is unknown. Therefore, we profiled the molecular landscape of CCNE1-amplified (amp) esophagogastric cancer (EGC) and explored treatment response outcomes. Methods: Tumors were analyzed by next-generation sequencing (NGS) of DNA (592 genes, NextSeq; WES, NovaSeq) and RNA (WTS, NovaSeq) (Caris Life Sciences, Phoenix, AZ). Deficiency in mismatch repair or microsatellite instability (dMMR/MSI-H) was tested by fragment analysis, IHC, and NGS. Tumor mutational burden (TMB) was measured by totaling somatic mutations per tumor (high>10 mt/MB). PD-L1 was tested by IHC (22C3, 1+). Real world overall survival (OS) was extracted from insurance claims and calculated from first of treatment to last contact using Kaplan-Meier survival curves for molecularly defined cohorts. Statistical significance was determined using chi-square and Mann-Whitney U test with p-values adjusted for multiple comparisons (q<0.05). Results: CCNE1Amp (copy number, CN ≥6) was identified in 142/2276 (6.2%) esophageal adenocarcinoma (EA), 101/1449 (7.0%) esophagogastric junction carcinoma (EJC), 109/2607 (4.2%) GA, and 6/751 (0.8%) esophageal squamous cell carcinoma (ESCC) samples. Metastatic sites such as lymph node and liver showed higher frequency of CCNE1Amp compared to primary sites. TMB, dMMR/MSI-H, and PD-L1 expression were similar in CCNE1-amp vs non-amp tumors. CCNE1Amp was enriched in CIN type tumors (defined by increased TP53 and ERBB2 amp) and was associated with decreased mutation of CDKN2A and CDH1. In addition, CCNE1-amp tumors showed increased HER2 positivity/ ERBB2 amplification by IHC (14.3% vs 10.1%) or CN analysis (19.1% vs 10.3%). Differential gene expression analysis demonstrated increased liver X receptor/retinoid X receptor (LXR/RXR) pathway activation in tumors harboring CCNE1Amp. While there were no differences in OS between CCNE1-amp vs non-amp GA, CCNE1Amp was associated with worse OS after trastuzumab in the IHC HER2+ cohort (N=9 vs 28; HR (95% CI): 2.95 (1.18 – 7.34), p = 0.015). By contrast, there was a trend toward improved OS after immunotherapy in CCNE1-amp GA (N=9 vs 194; HR (95% CI): 0.34 (0.11 –1.07), p = 0.054). Conclusions: CCNE1Amp is associated with a distinct molecular profile in EGC and resistance to HER2-targeted therapy. While CCNE1-amp tumors are thought to be generally “immune-cold,” CCNE1-amp GA demonstrated potentially improved outcomes with immunotherapy. Further investigation of CCNE1Amp as a predictive biomarker is warranted, particularly as novel therapeutics selectively targeting CCNE1Amp are under clinical investigation.

Tumor mutation burden in colorectal cancers with POLE exonuclease and non-exonuclease domain variants

224

Background: Driver mutations in the exonuclease domain (ExoD) of DNA polymerase epsilon ( POLE) have been associated with high tumor mutation burden (TMB). High TMB is an important marker for immunotherapy, and thus understanding the mechanisms that lead to high TMB is an area of active investigation. Many tumors with POLE ExoD driver mutations have additional POLE variants of uncertain significance (VUS). Some of these VUS lie within the ExoD, while most lie outside the ExoD. This study investigates the role of such POLE VUS on TMB and POLE functionality of ExoD-mutated tumors. Methods: Retrospective colorectal cancer (CRC) genomic profiles from Caris Life Sciences (n=1,870) were separated into groups based on the presence of POLE ExoD mutations and the TMB status. Groups were defined as: ‘TMB-high (TMB-H) POLE ExoD driver without additional POLE VUS’, ‘TMB-H POLE ExoD driver plus POLE VUS’, and ‘TMB-low (TMB-L) POLE variant(s) without ExoD driver’. An additional group with ‘TMB-H POLE variant(s) without ExoD driver’ was identified as potentially novel drivers for future study. The microsatellite instability status (MSI or MSS) was also considered. The level of TMB, the presence of polymerase proofreading-associated mutational signatures and POLE stability/functionality analyzed using AlphaFold2 models and Rosetta Molecular Modeling Suite, were evaluated. Results: POLE variants were identified in 4.9% of the CRCs studied (92/1870). In all, 52.2% (48/92) of those tumors were TMB-H, and 38.0% (35/92) of them contained POLE ExoD drivers. The highest median TMB (mTMB) was found in the ‘TMB-H POLE ExoD driver plus POLE VUS’ group, and this was significantly different from the ‘TMB-H POLE ExoD driver without additional POLE VUS’ and the ‘TMB-L POLE variant without ExoD driver’ group, even when MSI tumors were excluded (p<0.001, Mann-Whitney). These results were validated in TCGA CRC tumors and in other tumor types. Within the ‘TMB-H POLE ExoD driver plus POLE VUS’ group, the mTMB increased with the number of accumulated POLE VUS (p<0.001, Mann Whitney). The nucleotide sequence context of 77.8% of the POLE VUS matched mutational spectra associated with POLE ExoD defects, suggesting that most of the POLE VUS are secondary to the ExoD driver mutation. Finally, the majority of POLE VUS within ‘TMB-H POLE ExoD driver plus POLE VUS’ group have a predicted effect on POLE stability and/or function. Conclusions: Our results suggest that the presence of secondary POLE non-driver variants increase TMB in POLE ExoD-mutated tumors. Although still preliminary, these data could impact prognosis and response to immune checkpoint inhibitors in patients. [Table: see text]

Age- and sex-based differences in the genomic profiles of patients with gastrointestinal (GI) and pancreatic neuroendocrine neoplasms (NENs)

655

Background: Demographic variables such as age and sex have long been associated with risk of cancer incidence, therapeutic response, and mortality. In NENs, specifically, studies have shown differences in overall survival based on age at diagnosis and sex. However, studies investigating the molecular determinants associated with age and gender have been lacking. To delineate the mechanisms of age, gender, and outcomes, we queried a large real-world dataset of NENs originating from the GI-tract and pancreas and characterized their molecular and immune profiles. Methods: 1935 cases (GI: n=1431, Pancreas: n=504) of NENs were analyzed using Next Generation Sequencing or Whole Exome Sequencing of DNA, and 1211/1935 cases (GI: n=917, Pancreas: n=294) underwent Whole Transcriptome Sequencing at Caris Life Sciences (Phoenix, AZ). To study the age-associated effect on the molecular and immune landscapes, we applied two age cutoffs: below 50 and 70 years of age at the time of tissue collection, respectively. Significance was determined using chi-square, Fisher-Exact or Mann-Whitney U, and p-values were adjusted for multiple comparisons (q<0.05). Results: In GI-NEN, TP53, RB1 mutations and high tumor mutation burden (TMB_H; >=10 Mut/Mb) were more prevalent and KRAS mutations were less prevalent with increasing age. Interestingly, in pancreatic NENs (P-NENs), KRAS mutations were more prevalent and TMB_H were less prevalent with increasing age. Mutations specific to the sites of origin such as APC (GI) and MUTYH (pancreas) were negatively associated with age while SMAD4 (pancreas) was positively associated with age. While no differences in immune checkpoint gene (ICG) expression were associated with age, immune cell estimates of M1 macrophages (anti-tumorigenic) in GI-NENs were higher in patients above 70 and M2 macrophages (pro-tumorigenic) were higher in patients below 50 (p<0.05, q>0.05). For the analysis with gender, mutations in p53 were more prevalent among males with GI-NEN while dMMR/MSI-H were more prevalent in females with P-NENs. No significant differences in ICG expression and immune cell estimates were associated with gender. Conclusions: Our study queries one of the largest datasets of GI and P-NENs to date. We observe distinct molecular profiles associated with age and gender in both GI- and P-NENs as well as a distinct immune profile associated with age in GI-NENs. While these results are only hypothesis generating, they are suggestive of how patient demographics may be utilized for rational therapy design. [Table: see text]

Beyond CPS for PD-L1 scoring: Genetic alterations that impact efficacy of immunotherapy in hepatocellular carcinoma (HCC)

592

Background: Tumor-agnostic approvals of immune checkpoint inhibitors (ICI) include deficient mismatch repair/microsatellite instability high (dMMR/MSI-H) and high tumor mutational burden (TMB) while in cancer like metastatic gastroesophageal cancers, ICI use has been predicated on PD-L1 expression. ICI are increasingly used for metastatic HCC, but without required markers. We aimed to examine the genomic landscape of HCC in context of PD-L1 expression, and to determine clinical responses to ICI in this setting. Methods: Next-generation sequencing of DNA (592 or WES) and RNA (WTS) was tested at Caris Life Sciences (Phoenix, AZ). PD-L1 expression was tested by IHC (SP142) and compared as high (2+,5%), low (1-2,1%) and negative (0). dMMR/MSI-H was tested by IHC/NGS and TMB-High was defined as ³10 mutations/MB. QuantiSEQ was used to estimate the tumor microenvironment. X2/Fisher-Exact were used and significance was determined as P-value adjusted for multiple comparison ( Q < 0.05). Real-world overall survival (rwOS) was obtained from insurance claims and calculated from tissue collection to last contact; time-on-treatment (TOT) was calculated from start to finish of ICI treatments. Results: Overall, 17.7% of HCC expressed PD-L1 by IHC; 79/1306 (6.1%) had high expression. The overall prevalence of dMMR/MSI-H was 0.2%; TMB was high in 5.1%. PD-L1 expression was not associated with MSI-H or high TMB. When comparing tumors that are PD-L1 high vs. negative vs. low, expression of several immuno-oncologic (IO) markers LAG3 (median TPM: 2.4, 0.8, 0.5), CTLA-4 (2.9, 1.2, 0.5), IDO1 (4.2, 1.3, 0.6) and others, as well as T-cell inflamed and IFNg scores all decreased significantly with PD-L1 (all q<0.05), similar trends were seen with B cells, M1 macrophages, CD8+ T cells and T regs while opposite differences seen in NK cells (q<0.05). Additionally, when PDL1-high tumors were compared to PDL1-negative, mutations in CTNNB1 trended lower (21% vs 35%) while amplifications of KRAS (4% vs 0%), PDCD1LG2 (4% vs 0%) and mutations in HNF1A (3% vs 0%), HOXB13 (6% vs 0%), STK11 (5% vs 0%) trended higher; when PDL1-low were compared to PDL1-negative, mutations in TP53 (45% vs 32%), ELF3 (3% vs 0%), TSC2 (6% vs 2%) and PIK3CA (5% vs 1%) and amplifications in BCL9 (2% vs 0%) and CCNE1 (2% vs 0%) trended higher. When investigating clinical outcome of a cohort of 908 HCC tumors, PD-L1 expression had no effect on prognosis, and was not associated with differences in TOT of ICI in HCC. Conclusions: Prevalence of dMMR/MSI-H and TMB-H is very low in HCC. PD-L1 is only expressed in <20%. Even with a finding of strong association of expression of several established IO biomarkers with PD-L1 expression, it’s still not predictive of response to ICI. New biomarkers or molecular signatures need to be identified and validated to objectively identify HCC patients most likely to benefit from ICI treatment.

Comparative analysis of the molecular profile and tumor immune microenvironment (TIME) of human epidermal growth factor receptor 2 (HER2) low (L)- versus high (H)-expressing gastroesophageal cancers (GEC)

287

Background: Addition of immune checkpoint blockade to anti-HER2 therapy has improved outcomes in HER2-positive GEC. Anti-HER2 antibody-drug conjugates have shown activity in some HER2-L tumors in other tumor types. We aimed to compare the molecular profile and TIME of HER2-L and HER-H GEC. Methods: 8678 GEC (gastric, GE junction, and esophageal) adenocarcinoma and squamous cell carcinoma samples were analyzed by next-generation sequencing (NGS) of RNA (whole transcriptome, NovaSeq), DNA (592 genes, NextSeq, or whole exome sequencing, NovaSeq), and immunohistochemistry (IHC, Caris Life Sciences, Phoenix, AZ). Cohorts were stratified by IHC HER2 values of 0 (non-expressors), 1-2+ (HER2-L), or 3+ (HER2-H) and compared using X2 or Fisher-Exact. Statistical significance was determined as P-value adjusted for multiple comparisons (q < 0.05). Microenvironment cell population (MCP) counter was used to quantify immune cell infiltration. Results: Tumor were grouped into HER2 non-expressors (N = 5217), L (N = 2660), and H (N = 801). Mutations of TP53 (72% vs 92%) and amplification of MYC (4% vs 7%), CCNE1 (7% vs 12%), CCND3 (2% vs 4%), CDK6 (3% vs 6%), SMARCE1 (2% vs 25%) and RARA (3% vs 24%) were significantly lower in HER2-L compared to HER2-H (q < 0.05). ARID1A (14% vs 9%), PIK3CA (8% vs 3%), KRAS (9% vs 2%), GNAS (2% vs 0.3%), KMT2D (6% vs 1%), CDH1 (5% vs 1%), and ATM (4% vs 1%) mutations were significantly higher in HER2-L compared to HER2-H (q < 0.05). HER2-L was associated with more TMB-H (9.3% vs. 5%; q<0.05), MSI-H (4.7% vs. 0.6%; q<0.05), and a trend towards higher PD-L1 expression than in HER2-H (80.8% vs 76.1%, p < 0.05 but q > 0.05). Immuno-oncology (IO)-related gene expression inversely correlated with HER2 expression with lowest expression of PDCD1LG2, CD274, CTLA4, PDCD1, HAVCR2, CD80, IFNG, LAG3, and CD86 in HER2-H (q < 0.05). HER2-L had significantly higher median immune infiltration of B cells (fold change [FC]: 1.22), T cells (FC: 1.16), CD8+ T cells (FC: 1.56), NK cells (FC: 1.12), neutrophils (FC: 1.10), cytotoxic lymphocytes (FC: 1.36), and myeloid dendritic cells (FC: 1.34), compared to HER2-H (q < 0.05). HER2 non-expressors showed similar immune cell infiltrates compared to HER2-H. HER2-H was associated with lower T-cell inflamed scores and IFN gamma signature when compared to HER2-L and non-expressors (q < 0.05). Conclusions: To our knowledge, this study is the first and largest comparison of molecular profile and TIME in HER2-expressing GEC. We demonstrated distinct molecular and TIME profiles with higher immunogenic profiles in HER-L as compared to HER2-H. IO-related gene expression and TIME cell distribution differences in HER2-H GEC suggest that response to IO and HER2 therapy combinations is likely related to HER2-targeted treatment effect on TIME rather than baseline immunogenicity of the tumor.

Molecular correlates of DSCR1 expression in colorectal cancer (CRC)

185

Background: Down syndrome (DS), a genetic disorder caused by trisomy of chr 21, is associated with a considerably lower risk for solid tumors and other angiogenesis related diseases. DSCR1 belongs to a family of evolutionary conserved protein-coding genes located on chr 21 and is highly upregulated in DS patients. Its product, calcipressin-1, has been shown to reduce cancer risk by suppressing angiogenesis. We previously reported that a germline polymorphism in DSCR1 was associated with time to recurrence in resected CRC and anti-VEGF treatment outcomes in metastatic CRC. Here, we analyzed the molecular landscape of CRC according to DSCR1 expression levels. Methods: 20,237 samples from CRC tested at Caris Life Sciences (Phoenix, AZ) with WTS (Illumina NovaSeq) and NextGen DNA sequencing (NextSeq, 592 Genes and NovaSEQ, WES) were analyzed. Top quartile transcripts per million (TPM) for DSCR1 expression were considered high (Q4) while bottom quartile low (Q1). Cell infiltration (CI) in the tumor microenvironment (TME) was estimated by RNA deconvolution analysis using QuantiSEQ. Interferon-gamma and T-cell inflamed signatures were also calculated from RNA data. X2 and Fisher-Exact tests were used and statistical significance was determined as P-value adjusted for multiple comparisons ( q < 0.05). Results: DSCR1 expression was higher in primary tumors than metastases (10.5 vs 8.1 median TPM, q < 0.05). No significant difference was observed in right- versus left-sided tumors, however rectal tumors showed the highest DSCR1 expression ( P < 0.05). Overall, high DSCR1 TPM were associated with TMB-high (11.3% vs 9.3%), dMMR/MSI-H (7.9% vs 5.9%), and PD-L1 (4.7% vs 3.1%) ( q < 0.01); the association with TMB-H was not significant in pMMR/MSS. DSCR1 high was associated with lower mutation rates of APC, KRAS, TP53 and amplification of FLT1/ FLT3, while higher mutation rates of KMT2C/D, BRAF, PTEN, RNF43, and RSPO3 fusions ( q < 0.0001). Gene set enrichment analysis showed that high DSCR1 expressing tumors were enriched in alterations of several pathways including hypoxia, apoptosis, DNA repair, KRAS signaling, inflammatory response, and oxidative stress-related pathways (all P < 0.05, FDR < 0.25). B cells, macrophages, neutrophils, NK cells, Tregs, cancer associated fibroblasts and endothelial cells were more abundant in the TME of tumors with high DSCR1 while myeloid dendritic cells were lower, regardless of MMR status (all q < 0.001). DSCR1 expression was associated with a higher T-cell inflamed signature and IFN score ( q < 0.05). Conclusions: This is the first and most extensive profiling study to investigate DSCR1 expression in CRC. Our data show a strong association between tumor DSCR1 gene expression and distinct molecular features and TME cell infiltration. These findings suggest that DSCR1 holds potential as a novel therapeutic target for CRC and may be an important player in TME modulation.

Tissue factor expression in colorectal cancer

250

Background: Tissue factor (TF), a component of the coagulation cascade, transmembrane receptor, and cofactor for factor VII/VIIa is expressed by subendothelial cells and critical to hemostasis, thrombosis, cell proliferation, angiogenesis, and metastasis. TF is suggested to have a critical role in colorectal cancer (CRC). Here we assessed the clinico-molecular features associated with TF expression (exp) in CRC. Methods: Tumor molecular profiling was performed for 14,0786 samples by NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) and immunohistochemistry (IHC) at Caris Life Sciences (Phoenix, AZ). TF-high (TF-H) and TF-low (TF-L) RNA exp were defined as ≥ 75th- and < 25th-percetile of TF transcripts per million (TPM), respectively. X2/Fisher-exact and Mann Whitney U tests were used for comparison, and P-values were adjusted for multiple comparisons. Cell infiltration in the tumor microenvironment (TME) was estimated by quanTIseq. Real-world overall survival (OS) information was obtained from insurance claims data and Kaplan-Meier estimates were calculated for molecularly defined patients (pts). Results: TF exp was higher in primary tumors compared to metastatic sites (2.3-fold, P < 0.001); higher in lung vs. liver metastases (0.6-fold and 0.3-fold, p < 0.001); higher in rectal (1.4-fold), right-sided (1.3-fold), and transverse (1.2-fold) compared to left-sided (P < 0.001); and highest in CMS1 tumors (CMS2 0.4-fold, CMS3 0.7-fold, CMS4 0.8-fold, P < 0.001). TF exp was higher in tumors with high TMB (TMB ≥ 10 Mut/Mb) (1.6-fold), dMMR/MSI-H (1.9-fold), and PD-L1 exp (1.8-fold) (P < 0.001 each). Compared to WT tumors, mutations in KRAS (1.2-fold, P < 0.001), BRAF (1.5-fold, P < 0.001) and ERBB2 (1.2-fold, P < 0.01) were associated with higher TF exp. In pMMR/MSS CRC, TF-L tumors were associated with TP53, APC, and NRAS mutations, and TF-H was associated with KRAS, PIK3CA, SMAD4, FBXW7, BRAF, ARID1A, GNAS, SMAD2, RNF43, KMT2D and BRCA1 mutations. TF exp was positively associated with most immune cell populations, most strongly with endothelial cells. In pMMR/MSS KRAS WT CRC, TF-H was associated with worse OS than TF-L. TF-H was also associated with worse OS for pMMR/MSS CRC pts treated with FOLFOX (HR: 1.5, CI:1.2-1.8, p < 0.0001), FOLFIRI (HR:1.7, CI:1.3-2.1, p < 0.0001), VEGF (HR:1.3, CI:1.1-1.6, p = 0.002) and EGFR inhibitors (HR:1.6, CI:1.1-2.3, p = 0.009) than TF-L. TF-H was associated with greater median days on pembrolizumab compared to TF-L in dMMR/MSI-H CRC (HR: 0.6, CI: 0.4-0.9, p = 0.018). Conclusions: Our data show TF exp is associated with distinct clinical and molecular features, including tumor sidedness and metastatic pattern, as well as CMS subtype and KRAS, BRAF and ERBB2 mutations. TME cell infiltration and IO-related biomarkers were enriched in TF-H, with TF exp correlating with endothelial cell abundance. These findings and associations with patient outcomes suggest that TF may be a relevant biomarker and target in CRC.

Analysis of concordance between microsatellite instability by next generation sequencing (NGS-MSI) and mismatch repair deficiency by immunohistochemistry (IHC-MMR) in >28,000 colorectal tumors

30

Background: The use of immune checkpoint inhibitors (ICI) in MSI-H/MMRd CRC patients has profoundly changed the treatment landscape in the metastatic setting; the most recent NICHE-2 trial showed 100% response rate in MSI-H/MMRd locally advanced CRC patients treated with neoadjuvant ICI, further highlighting the importance of identifying these patients accurately to enhance patient care. Here we report the concordance of NGS-MSI and IHC-MMR from a very large cohort of CRC tumors and study the molecular characteristics and clinical outcomes of these patients. Methods: A total of 28,105 CRC tumors were analyzed by NGS (592 genes, NextSeq or WES, NovaSeq) and IHC. MMRd was defined as complete loss of ≥1 IHC stains (MLH1, MSH2, MSH6, or PMS2) and proficient (MMRp) as any positive staining for all four proteins. MSI status was determined from over 7,000 target microsatellite loci covered by NGS. Central pathology review (CPR) of MMR IHC was done on 73 cases with discordant MMR/MSI results. Real-world overall survival was obtained from insurance claims and calculated from either tissue collection or treatment start to last contact; Kaplan-Meier estimates were calculated for molecularly defined patients. Results: For CRC tumors with NGS and IHC, 28031/28105 were concordant (99.74%), 0.09% were MMRd/MSS, and 0.2% were MMRp/MSI-H. After CPR, 23/24 (96%) of MMRd/MSS cases and 46/49 (94%) of MMRp/MSI-H were confirmed. Of the 46 post-CPR MMRp/MSI-H CRC samples, 52% had ≥1 pathogenic missense mutation in including MLH1/PMS2/MSH2/6 or MLH3, MSH3, PMS1 or POLE. Of the 23 post-CPR MMRd/MSS CRC samples, 38% had MLH1 loss, 91% PMS2 loss, no MSH2 loss and 9% MSH6 loss. When comparing clinical outcomes with concordant MSS/MMRp tumors, concordant MSI-H/MMRd patients had significantly longer OS (HR=1.131 [95% CI: 1.02-1.254], p<0.001) and post-ICI survival (HR= 2.695 [95% CI: 1.932-3.76], p<0.001). Compared to MSS/MMRd tumors (N=21), MSI-H/MMRp (N=29) trends to have longer OS (HR=2.163, [95% CI: 0.939-4.983], p=0.064) and insufficient ICI-treated patients were available for analysis. Conclusions: Here we report from >28,000 CRC tumors that the concordance of IHC-MMR/NGS-MSI is 99.74%. Clinical outcome from a large cohort of patients show NGS is not inferior compared to IHC in identifying patients with MSI-H/MMRd. The additional lens that NGS-MSI offers is of value in identifying CRC patients who may benefit from ICI therapy. [Table: see text]

The role of gene expression of CD47 in colorectal cancer (CRC)

240

Background: CD47 belongs to the immunoglobulin superfamily and is overexpressed in many tumor types. CD47 plays an important role in suppressing phagocytosis through binding to transmembrane protein SIRP-alpha on macrophages. Targeting CD47 is a novel strategy for cancer immunotherapy and is being evaluated in ongoing clinical trials. However, molecular characteristics of CD47-overexpressed colorectal cancer (CRC) are largely unknown. Methods: We retrospectively reviewed CRC patient samples (n = 14786) submitted to a commercial CLIA-certified laboratory (Caris Life Sciences, Phoenix AZ). Next-generation sequencing of DNA and RNA (whole-transcriptome sequencing) and immunohistochemistry were performed. Correlation of CD47 expression with danger-associated molecular pattern (DAMP)-related genes ( HMGB1, CALR, ANXA1, HSP90AA1, HSPA1A, and CXCL10) expressions was tested. DAMP signature calculated as composite z-score of the DAMP related genes was compared between CD47-high and -low patients classified according to the median level of CD47 expression. Distributions of KRAS and BRAF mutations, consensus molecular subtype (CMS), and signatures of oncogenic signaling pathways were compared between CD47-high and -low patients. In addition, overall survival (OS) was compared between CD47-high and -low patients available for survival data. Results: CD47 expression level was significantly higher in metastatic compared to primary lesions (1.07-fold, q < 0.05) and microsatellite instability high tumors compared to microsatellite stable tumors (1.15-fold, q < 0.05). CD47 expression was positively correlated with DAMP genes expression except for HSPA1A, and the DAMP signature (median score [MS]: 2.66 vs -2.29, q < 0.05) was significantly increased in CD47-high patients. KRAS mutations were less prevalent (45.87% vs 50.05%) and CMS1(17.72% vs 14.42%) and CMS4 (40.33% vs 27.28%) were more prevalent in CD47-high patients, while no difference was observed in the prevalence of BRAF mutations between CD47-high and -low patients. Signatures of EMT (MS:2.91 vs -3.19), TGF-beta (MS:3.72 vs -3.52), angiogenesis (MS:2.95 vs -2.78), MAPK (MS:4.99 vs -4.12), PI3K (MS:2.41 vs -2.03), and immune-related signaling pathways (MS:1.20 vs -2.51) were significantly enriched in CD47-high patients (all q < 0.05). CD47-high patients (n = 4873) showed significantly worse OS than CD47-low patients (n = 4898) (median OS, 32.4 vs 37.6 months; hazard ratio = 1.158, p< 0.01). Conclusions: Highly CD47-expressed CRC harbored activation of DAMPs and oncogenic signaling pathways that linked to aberrant tumor microenvironment and worse prognosis. Our results support intensive treatment strategies using CD47 inhibitors combined with cytotoxic agents and molecular targeted agents (such as anti-VEGF agents) in CD47-overexpressed CRC.

Survival of patients with colorectal cancer (CRC) with low expression of homologous recombination proficient (HRP) genes

225

Background: HR deficient (HRD) CRC has improved outcomes following exposure to DNA damaging agent (DDA) (irinotecan, IR; oxaliplatin, OX) compared to HRP CRC. Low expression of wild type (WT) BRCA1 mRNA is associated with prolonged OS in ovarian cancer; however, this finding has not been investigated in CRC or outside of BRCA. Here, we examine the effect of low expression of HR genes in HRP CRC on post-DDA survival. Methods: 12,860 CRC samples were analyzed by NGS (592, NextSeq; WES, NovaSeq) and WTS (NovaSeq) at Caris Life Sciences (Phoenix, AZ). Samples were classified by RNA expression percentiles. Real world OS was extracted from insurance claims and calculated using Kaplan-Meier estimates for molecularly defined cohorts from first of OX or IR to last contact. Results: Post-IR survival was prolonged with low expression of ATM, CHEK2 and PALB2 in WT ATM, PALB2, and CHEK2 WT CRC, respectively (bottom vs. top 10%, bottom vs. top 25%; p<0.05). Notably, low PALB2 expression (bottom 10%) showed a 14.5-month post-IR benefit compared to high PALB2 expression (top 10%) in WT PALB2 CRC (p=0.003). Post-OX survival was not significantly prolonged with low expression of CHEK2 in WT CHEK2 CRC (bottom 10% vs top 10%, bottom 25% vs top 25%) but was with low expression of ATM (+9.7 months, p=0.02) and PALB2 (+5.6 months, p=0.03) in WT ATM and PALB2 CRC, respectively (all bottom 25% vs top 25%). Conclusions: Here we report the first findings to suggest a novel subclass of CRC defined as the low expression of mRNA of non-mutated HRD genes that exhibit sensitivity to DDA. Low expression of WT ATM, CHEK2, and PALB2 correlates with prolonged OS following IR, post-OX survival was prolonged with low expression of ATM and PALB2. Further characterization defining sensitivity of low expressing HRP genes to DDA may help guide treatment considerations in HRP CRC. [Table: see text]

ERBB family fusions are recurrent and actionable oncogenic targets across cancer types

Purpose

Gene fusions involving receptor tyrosine kinases (RTKs) define an important class of genomic alterations with many successful targeted therapies now approved for ALK, ROS1, RET and NTRK gene fusions. Fusions involving the ERBB family of RTKs have been sporadically reported, but their frequency has not yet been comprehensively analyzed and functional characterization is lacking on many types of ERBB fusions.

Materials and methods

We analyzed tumor samples submitted to Caris Life Sciences (n=64,354), as well as the TCGA (n=10,967), MSK IMPACT (n=10,945) and AACR GENIE (n=96,324) databases for evidence of EGFR, ERBB2 and ERBB4 gene fusions. We also expressed several novel fusions in cancer cell lines and analyzed their response to EGFR and HER2 tyrosine kinase inhibitors (TKIs).

Results

In total, we identified 1,251 ERBB family fusions, representing an incidence of approximately 0.7% across all cancer types. EGFR, ERBB2, and ERBB4 fusions were most frequently found in glioblastoma, breast cancer and ovarian cancer, respectively. We modeled two novel types of EGFR and ERBB2 fusions, one with a tethered kinase domain and the other with a tethered adapter protein. Specifically, we expressed EGFR-ERBB4, EGFR-SHC1, ERBB2-GRB7 and ERBB2-SHC1, in cancer cell lines and demonstrated that they are oncogenic, regulate downstream signaling and are sensitive to small molecule inhibition with EGFR and HER2 TKIs.

Conclusions

We found that ERBB fusions are recurrent mutations that occur across multiple cancer types. We also establish that adapter-tethered and kinase-tethered fusions are oncogenic and can be inhibited with EGFR or HER2 inhibitors. We further propose a nomenclature system to categorize these fusions into several functional classes.

Abstract 6130: Comprehensive genomic and transcriptomic profiling of acral lentiginous melanoma

Background: Acral lentiginous melanoma (ALM) is a rare melanoma subtype found on the palms, soles and nailbeds. Outcomes are poor for patients with advanced ALM, and novel treatment approaches are needed. Here, we seek to explore the global genomic and transcriptomic landscape of ALM.

Methods: A total of 699 primary CM (non-ALM cutaneous melanoma) and 18 primary ALM samples underwent next generation sequencing of DNA (592 Gene Panel, NextSeq, or WES, NovaSeq), and whole transcriptome sequencing (NovaSeq, WTS). Wilcoxon, Fisher’s exact test were used to determine statistical significance (displayed as p value without and q value with multi comparison correction). xCell, HLA subtyping, neoantigen load (HBA: high binding affinity; IBA: intermediate binding affinity; LBA: low binding affinity), Interferon gamma score (IFNγ), MAPK pathway activity score (MPAS), and Innate anti-PD-1 Resistance score (IPRES) were calculated by mRNA expression. Global differentially regulated genes were assessed via limma R package (C: log fold change).

Results: The most common alterations in ALM included NRAS (22.2%), NF1 (20.0%), BRAF (11.1%) and CDKN2A (11.1%) mutations, and EMSY (22.2%), ELL (11.1%), MAML2 (11.1%), MRE11(11.1%) and PIK3R2 (11.1%) amplifications. ALM had lower TMB (1.5 v 9 Mut/Mb, q&lt;.0001), lower rates of TERT (0 v 66.7%, q&lt;.01) and a trend towards lower BRAF (11.1 v 39.8%, p&lt;.05) mutations, compared to CM. Neoantigen load was lower in ALM compared to CM, regardless of MHC binding affinity (HBA: 1 vs 4, q &lt;.01; IBA: 2 vs 7, q &lt;.001; LBA: 7 vs 18, q &lt;.001). HLA-G RNA expression was upregulated in ALM with respect to CM (C = 1.14, q &lt;.001). ALM showed less CD4+ T cell Th1 (C = -0.8, p &lt;.05), B cell plasma (C = -1.8, p&lt;.05), and γδ T cells (C = -5.9, p&lt;.05), but more CD4+ T cell central memory cell (C = 9.7, p&lt;.05), stroma score (C = 1.7, p&lt;.05), and endothelial cells (C = 1.7, p&lt;.05), versus CM. There was a trend towards lower IFNγ in ALM (-0.4 vs -0.3, p = .1), but no difference in IPRES, compared to CM (-0.09 vs 0.1, p=.9). MPAS scores were lower for ALM compared to CM (-1.6 vs -0.4, q&lt;.001), even when stratifying by BRAF (q&lt;.05) or NF1 (q&lt;.05) status, but not NRAS (p = .22). Pathways related to keratinization (p &lt;.0001) and amyloid fiber formation (p&lt;.0001) were enriched in ALM, due to overexpression of KRT16 (C = 3.5, q &lt;.01), KRT6B (C = 3.4, q &lt;.01), and KRT17 (C = 3.2, q &lt;.05), among others.

Conclusion: ALM has distinct immunologic features, including upregulation of HLA-G, as well as lower MAPK activation in ALM, compared to CM, highlighting the need for novel therapeutic approaches in the treatment of this rare subtype.

Citation Format: Gino K. In, Jun Yin, Phillip Walker, Justin Moser, Joanne Xiu, Kelsey Poorman, Geoffery T. Gibney, Matthew Oberley, Thuy Phung, Leonel F. Hernandez-Aya, Jose Lutzky, Wolfgang Michael Korn, Michael B. Atkins. Comprehensive genomic and transcriptomic profiling of acral lentiginous melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6130.

Abstract 5699: Overexpression of KMT2A is associated with worse prognosis and specific immune signatures in patients with TP53-mutated hepatocellular carcinomas

Background: Aberrant expression of epigenetic regulators is often associated with pathogenesis. Histone H3K4 methyltransferase, known as KMT2A, has been implicated in regulation of chromosome segregation, mitosis and DNA replication in pediatric leukemia and myeloma. However, the role of KMT2A expression in solid tumors is under-investigated. Here we examine the implications of KMT2A overexpression in prognosis, gene pathway enrichment, aneuploidy and immune infiltration patterns using a large, real-world clinical HCC dataset.

Methods: A total of 403 HCC samples underwent comprehensive molecular profiling at Caris Life Sciences, including DNA-(592 Gene Panel, NextSeq, or whole exome sequencing, NovaSeq) and RNA- (NovaSeq, whole transcriptome sequencing, WTS) sequencing. Wilcoxon, Fisher’s exact test were used to determine statistical significance (p value without and q value with multi comparison correction). Aneuploidy scores were generated from CNVkit. Apoptotic index (AI), GSEA were assessed using mRNA levels (FDR&lt;0.25 as cutoff). Overall survival was calculated from date of tissue collection to date of last contact from insurance claims data and used for Kaplan-Meier method.

Results: Overexpression of KMT2A predicts poor survival in patients with HCC (HR 2.6, 95% CI [1.4 - 5.2], p&lt;.01). KMT2Ahigh HCC has higher TP53 (57.9% vs 28.8%, p&lt;.001) and lower CTNNB1 (14.1% vs 34.6%, p &lt;.01) mutation rates (mt). GSEA analysis showed that pathways such as mitotic spindle (NES = 2.1), DNA repair (NES = 2.0), E2F regulation (NES = 2.0) and MYC (NES = 1.9) are significantly enriched in KMT2Ahigh HCC. Interestingly, in KMT2Ahigh HCC, TP53 mutation status is a stratification factor for several HCC features: 1) TP53 mt HCC displayed significantly lower aneuploidy score (median 6 vs 13.5, p&lt;.01) and higher apoptotic index (median 1.1 vs 1.2, p&lt;.05); 2) inflammatory response pathways (NES = 2.1) and IL6 JAK STAT3 signaling (NES = 2.1) are specifically enriched in TP53 mt HCC and 3) TP53 mt HCC has more infiltration of B cells (5.3% vs 3.3%, q&lt;.05), Macrophage M1 (6.4% vs 2.9%, q&lt;.01), CD8 + T cells and myeloid dendritic cells (1.1% vs 1%, q&lt;.01) while TP53 wt is associated with more infiltration of Macrophage M2 (5.5% vs 3.3%, q&lt;.05) cells.

Conclusions: KMT2A could act as an independent prognostic marker in HCC. The negative correlation between KMT2A expression and aneuploidy scores in TP53 mt indicates potential roles of KMT2A in maintaining genome stability. Furthermore, our results suggest TP53 status is an important stratification factor for HCC with KMT2A overexpression. Our results warrant further investigation on the impact of KMT2A level on immune modulation and may define a subset of HCC that responds most effectively to immune checkpoint inhibition.

Citation Format: Liang Sha, Jun Yin, Sungming Kim, Woojin An, Jian Zhang, Alex Farrell, Joanne Xiu, David Spetzler, Shuanzeng Wei, Dave S. Hoon, Stephen V. Liu, Emil Lou, Misako Nagasaka, Wafik S. El-Deiry, Benedito A. Carneiro, Wolfgang Michael Korn, Heinz-Josef Lenz, Yali Dou. Overexpression of KMT2A is associated with worse prognosis and specific immune signatures in patients with TP53-mutated hepatocellular carcinomas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5699.

Abstract 5625: Molecular and immunologic characterization of HRAS mutations in a cohort of 6,329 patients with cutaneous melanoma

Background: Activation in RAS pathway has been associated with cancer development. Three RAS family members, including NRAS, KRAS and HRAS are frequently mutated across various cancer types, where NRAS mutations are present in 15-20% of melanomas. NRAS-mutant melanomas (NRASm) have been extensively characterized. However, molecular and clinical implications of HRAS mutations (HRASm) in melanoma are less well understood.

Methods: A total of 6329 melanoma samples were subjected to comprehensive molecular profiling at Caris Life Sciences. Analyses included next generation sequencing of DNA (592 Gene Panel, NextSeq; whole exome sequencing, NovaSEQ), RNA (NovaSeq, whole transcriptome sequencing, WTS) and IHC. MPAS scores to evaluate MAPK pathway activation, IFN scores, QuantiSeq, neoantigen load (high, intermediate, low binding affinity: HBA, IBA and LBA) and GSEA were calculated from mRNA expression data. Wilcoxon, Fisher’s exact were used to determined statistical significance (p value without and q value with multi comparison correction; FDR for GSEA). The reference cohort was the entire melanoma cohort (MC).

Results: HRASm were identified in 69 (1.09%) of melanoma samples (hotspots mutations: G13, 40%; Q61, 34%; G12, 18% and others, 9%). HRASm and NRASm had different genomic landscapes: HRASm were significantly associated with a higher mutation rate of NF1 (43.2% vs 27.7%, p&lt;.05), ARID1A (17.2% vs 6.3%, p&lt;.05), B2M (14.3% vs 2.4%, p&lt;.05), RAF1 (12.2% vs 1.4, p&lt;.0001), CTNNB1 (9.1% vs 3.3%, p&lt;.05) and higher amplifications of EMSY (11.8% vs 1.8%, p&lt;.01), MRE11 (4.3% vs 0.5%, p&lt;.05), whereas NRASm harbored less NF1 (14.8% vs 27.7%, q&lt;.0001 ), BRAF (6.9% vs 39.9%, p &lt;.0001), PTEN (3.9% vs 6.9%, q &lt;.05), KIT (0.8% vs 4.4%, q &lt;.0001) mutations and less amplification PDGFR (0.4% vs 1.3%, p&lt;.05), BRAF (0.2% vs 1.4%, p&lt;.05), KIT (0.2% vs 2.1%, q&lt;.05) when comparing to MC. Both HRASm and NRASm had higher MPAS scores than MC (HRASm, 0.24; NRASm, 0.11; MC, -0.41, q&lt;.001). In addition, HRASm showed higher TMB (HRASm, 68.1%; NRASm, 56.9%; MC, 50.0%, q&lt;.05), relatively higher IFN scores (HRASm, 0.16; NRASm, -0.23; MC, -0.24, q = .16) and higher neoantigen load (HBA: HRASm, 10.5; NRASm, 4; MC, 4, p &lt; .05; IBA: HRASm, 17.5; NRASm, 8 MC, 7, p &lt; .05; LBA: HRASm, 37.5; NRASm, 21; MC, 19, p =.1) when compared to NRASm and MC. Lastly, suppression of angiogenesis pathway was observed in both HRASm (NES = 1.7, FDR&lt;.05) and NRASm with respect to MC (NES = 1.4, FDR&lt;.25).

Conclusions: The genomic landscape of HRASm are significantly different from that of NRASm, implying their distinct roles in tumorigenesis. HRASm also demonstrated higher MAPK activation, suggesting that they could potentially benefit from agents targeting on this pathway. In addition, HRASm displayed more immunogenic features, associated with down-regulation of angiogenesis pathway, revealing a potential higher susceptibility of HRASm to immunotherapy.

Citation Format: Leonel F. Hernandez-Aya, Estelamari Rodriguez, Aparna Nallagangula, Jun Yin, Phillip Walker, Joanne Xiu, Justin Moser, Gino K. In, David Spetzler, Geoffery T. Gibney, Matthew Oberley, Thuy Phung, Michael Atkins, Dave S. Hoon, Wolfgang Michael Korn, Jose Lutzky, Gilberto Lopes. Molecular and immunologic characterization of HRAS mutations in a cohort of 6,329 patients with cutaneous melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5625.

The tumor microenvironment and immune infiltration landscape of KRAS mutant pancreatic ductal adenocarcinomas (PDAC) compared to colorectal adenocarcinomas (CRC)

4142

Background: The composition of the tumor microenvironment (TME) in PDACs is more heavily driven by mutant (mt) KRAS than any other cancer. How genomic alterations of PDAC including KRAS status affect the immune cell (IC) landscape remains unclear. Thus, we characterized IC types and the prevalence of immuno-oncologic (IO) biomarkers in PDAC by genomic and transcriptomic analysis, and investigated associations of mt KRAS with IC estimates in the TME. Our findings were compared to our previous study in CRC. Methods: A total of 4,142 PDAC and 3,727 CRC with KRAS- mts were analyzed using next-generation DNA sequencing (NextSeq, 592 gene panel or NovaSeq, WES), IHC, and whole transcriptome RNA sequencing (NovaSeq) (Caris Life Sciences, Phoenix, AZ). MSI/MMR was tested by FA, IHC and NGS. TMB-H was classified based on a cut-off of >10 mutations per MB. ICs were estimated by QuantiSeq (Finotello 2019, Genome Medicine) or MCP counter (Betcht 2016, Genome Biology). Significance was determined by X2 and Fisher-Exact and p-adjusted for multiple comparisons (q<0.05). Results: Mutant KRAS was seen in 81% of PDAC and in 48% of CRC. The most common variant was G12D, comprising 43% and 32% of all PDAC and CRC KRAS variants, respectively. The therapeutically actionable KRAS G12C variant comprises 2% and 7% of PDAC and CRC in this cohort, respectively. In PDAC, KRAS mt was associated with lower prevalence of MSI-H/dMMR than KRAS-wildtype (wt); 0.9% vs 1.9%, p=0.027). PDL1 expression was significantly lower in KRAS wt (12%) compared to G12D (19%) and G13X (33%), similar to previous observations in CRC. However, when considering TMB, in PDAC, G12D (1%), G12V (1%) and Q61 (1%) mutations had significantly lower TMB-H than RAS wt tumors (4%); in contradiction to CRC. The immune cell environment of KRAS mt PDAC showed significantly higher infiltration with M1 macrophages and cancer-associated fibroblasts (CAFs), as well as lower M2 macrophages, CD4+ & CD8+ T cells, T-reg, NK, myeloid dendritic and endothelial cells compared to KRAS wt. In CRC, a similar pattern was observed but more pronounced in PDAC. Immune-regulatory markers, were among multiple genes downregulated in KRAS-mt PDAC, including CTLA-4 and LAG3. Overall changes were most pronounced in cases harboring KRAS G12D, G12V, Q61, and rare KRAS variants. Conclusions: The TME of KRAS mt PDAC shows IC patterns similar to KRAS mt CRC. Actionable IO-targets, such as PDL1, are enriched in tumors harboring specific variants of KRAS mt PDAC including the targetable G12C variant. If G12D becomes druggable, it could be targetable in 35% patients with PDAC or 15% in CRC. These results demonstrate that the TME of PDAC and CRC shows immune-cold features. Tailored immunotherapeutic strategies would have to overcome these barriers in KRAS mt PDAC and CRC, possibly in combination with molecularly targeted treatment strategies.

Comprehensive genomic and transcriptomic characterization of small bowel adenocarcinoma

4018

Background: Small bowel adenocarcinoma (SBA) is a rare cancer with rising incidence and worse overall survival (OS) compared to other intestinal cancers. Clinical management of SBA is primarily extrapolated from colorectal cancer (CRC). Comprehensive genomic and transcriptomic profiling of SBA will facilitate development of disease-specific therapeutic strategies. We investigated molecular alterations and association with clinical outcomes in a large cohort of SBA pts. Methods: Tumors were analyzed using 592 gene next-generation sequencing of DNA (592 genes or WES), RNA (WTS) and IHC (Caris). Immune infiltration was calculated by QuantiSeq. OS was calculated from treatment start/tissue collection to last contact from insurance claims. Results: We analyzed 823 SBA tumors: 448 primary/312 metastases, 586 duodenal (DA)/95 jejunal (JA)/38 ileal (IA). Median age of JA pts was lower (57yo) than DA (67yo) and IA (68yo). Upon subsite comparison, HER2 overexpression (2.5%) and amplification (3.6%) was only seen in DA, while HER2 mutations (mts) were most common in JA (10%) and absent in IA. IA had the lowest rate of KRAS (30%) and APC (11%) mts and highest rate of DDR mts (37%). JAs were enriched in RSPO3 fusion (19%) and BRAF mts (21%). Among BRAF mts in SBA, class 3 comprised 53%, class 2 37% and class 1 10%. MSI/dMMR was seen in 8% SBA and TMB-H in 11% with no difference among subsites. Compared to 14000 CRC tumors, SBA had significantly higher immune infiltrates regardless of MSI status (p<0.001), with highest fold change in myeloid dendritic cells (15.5), Tregs (9.4), neutrophils (3.6) and M2 Macrophages (3.5). When investigating clinical outcome of SBA pts (n=751), favorable prognostic markers included TMB³8 mts/Mb (HR: 0.65, 95%CI: 0.50-0.85), mts in APC (HR: 0.76, 95% CI: 0.62-0.93), MSH6 (HR: 0.45, 95% CI: 0.20-0.99), HNF1A (HR: 0.268, 95% CI: 0.11-0.65), PRKDC (HR: 0.45, 95% CI: 0.20-1.01) and ERBB3 (HR: 0.46, 95% CI: 0.26-0.82), while TP53 (HR: 1.32, 95% CI: 1.10-1.59) and CDKN2A (HR: 1.7, 95% CI: 1.25-2.3) mts and positive PD-L1 (HR: 1.44, 95% CI: 1.09-1.90) predicted worse OS. Among SBA pts treated with chemotherapy (n=258), DA had worse OS than IA/JA (HR: 1.44, 95% CI: 1.05-1.98), which had worse OS than left-sided [LS] (HR: 1.86, 95% CI: 1.39-2.46) and right-sided [RS] CRC pts (HR: 1.35, 95% CI: 1.01-1.79). DA pts had significantly worse OS compared to LS (HR=2.97, 95% CI: 1.73-5.08) and RS (HR=1.88, 95% CI: 1.08-3.24) CRC pts. Conclusions: This study represents the largest SBA cohort with comprehensive genomic and transcriptomic profiling. We identified subsite-specific enrichment in targetable alterations, including HER2 overexpression/amplification in DA, BRAF/HER2 mts and RSPO3 fusions in JA, and DDR mts in IA. SBAs harbor higher immune infiltrates than CRC, suggesting active immune modulation. DA is characterized by poor overall outcomes and decreased therapeutic benefit from chemotherapy compared to LS- and RS- CRCs.

Impact of somatic SWI/SNF alterations on the genomic landscape of pancreatic ductal adenocarcinoma and response to PARP and immune checkpoint inhibitor therapy

e16289

Background: The SWI/SNF (SWItch/Sucrose Non-Fermentable) chromatin remodeling complex is an adenosine triphosphate (ATP)-dependent group of proteins that controls the accessibility of transcription factors to DNA and is a dynamic epigenetic regulator of gene expression. This system is deranged in about 14% of pancreatic ductal adenocarcinoma (PDAC) patients (pts). A recent report suggested that SWI/SNF alterations are associated with response to immune checkpoint inhibitors (ICIs). Herein, we investigate the relationship between SWI/SNF alterations and genomic instability, prognosis and response to PARP inhibitors and ICIs. Methods: Tumor molecular profiling was performed on PDAC FFPE samples by NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) at Caris Life Sciences (Phoenix, AZ). MSI/MMR status was determined by IHC, NGS and fragment analysis. Tumor mutational burden high (TMB-H) was determined by a cutoff of 10mut/MB, PD-L1 was tested by IHC (SP142) with a cutoff of 2+, 5%. SWI/SNF mutant tumors (MT) were compared to wild-type (WT), and significance was determined by X2 or Fisher-Exact and p adjusted for multiple comparisons (q) of < 0.05 (Benjamini-Hochberg correction). Real-world overall survival was obtained from insurance claims data and calculated from tissue collection to last contact; time on treatment (TOT) was calculated from treatment start to finish and compared by the Kaplan-Meier test. Results: Among 5075 PDAC tumors, 311 (6.1%) harbored SWI/SNF alterations (186 ARID1A-mutant, 60 PBRM1, 65 SMARCA4 and 12 SMARCB1). The proportion with SWI/SNF alterations in MSI-H tumors (23/53 or 43%) was higher than MSS (287/4922 or 5.8%, p < 0.001). Since MSI-H PDAC have distinct genomic behavior compared to MSS, MSS tumors were analyzed separately. TMB-H (8.3% vs. 1.2%) and PD-L1-H (23% vs. 15%) were more prevalent in MT than WT (q < 0.05) but not different in MSS tumors (TMB-H: 2.9% vs. 0.6%, q = 0.09; PD-L1: 21.4% vs. 15.4%, q = 0.3). Among the 2268 WES-tested tumors, gLOH-high was not different in MT vs WT (9% vs. 12%, q = 1) in all or in MSS tumors (9.7% vs 11.9%, q = 1). Among the 3728 pts with outcome data, MT pts had worse prognosis (HR = 0.78, 95% CI [0.70-0.87], p < 0.00001), similarly in MSS cohort (HR = 0.734, [ 0.66-0.82], p < 0.00001). Among 50 PARPi-treated pts, no TOT difference was seen between MT vs WT (HR = 0.913, [0.309-2.701], p = 0.9), similarly in MSS pts (HR = 1.104 [0.426–2.86], p = 0.9). In the 38 ICI-treated pts, no difference was seen in TOT between MT vs WT (HR = 1.42 [0.73-2.8], p = 0.3], similarly in MSS pts (HR 1.175 [0.524–2.636], p = 0.7). Conclusions: In this real-world database, SWI/SNF alterations were associated with a significantly worse prognosis. There was no association with increased genomic instability of PDAC tumors, and in the cases with linked treatment data, there was no association with response to PARPi’s or ICI’s.

The differential response to immune checkpoint inhibitors in colorectal and endometrial cancer patients according to different mismatch repair alterations

3625

Background: In colorectal cancer (CRC) and endometrial cancer (EC) patients (pts), preliminary data suggest a differential response to immune checkpoint inhibitors (ICIs) according to different MMR alterations. The drivers of this difference remain unknown and no reliable predictive biomarker has been found. We explored the genomic alterations, tumor mutation burden (TMB), immune-related gene expressions and signatures, tumor microenvironment (TME), neoantigen load and median overall survival (mOS) inCRC and EC pts treated with ICIs with different MMR alterations. Methods: 13,701 CRC and 3,315 EC specimens were tested at Caris Life Sciences (Phoenix, AZ) with Next Gen Sequencing (NGS) of DNA (592-gene or whole exome) and RNA (whole transcriptome). MMR/MSI status was determined by IHC of MMR protein and/or NGS. Immune cell abundance was quantified using quanTIseq. Gene expression profiles were analyzed for T cell-inflamed signature (TIS) and IFN-gamma scores. Immune epitope prediction was performed using the NetMHCpan v4.0 method in the Immune Epitope Database. Real-world mOS was obtained from insurance claims data and calculated from tissue collection or ICIs start to last contact. Statistical significance was determined using chi-square/Fisher-Exact and adjusted for multiple comparisons (adjusted p < 0.05). Results: In CRC, 84 (0.6%) pts had intact expression of MLH1 and PMS2 and co-loss of MSH2 and MSH6 (MutS) and 648 (4.7%) had co-loss of MLH1 and PMS2 and intact MSH2 and MSH6 (MutL). 117 (0.9%) had other MMR IHC loss. APC, KRAS, ERBB2, ERBB3 and MSH2 mutations rates were higher in MutS while BRAF mutation rate was higher in MutL. B cell, NK cell content and neoantigen load (high affinity epitopes: p < 0.05, intermediate: p < 0.01, low: p < 0.001) were higher in MutS. The mOS in MutS (N = 149) vs. MutL (N = 980) was 56 months (m) vs. 36 m (p = 0.003). In ICI-treated pts, the mOS in MutS (N = 28) vs. MutL (N = 149) was not reached (NR) vs. 32 m (p = 0.005). BRAF mutation didn’t impact survival in MutL. In EC, 48 (1.4%) pts had MutS and 915 (27.6%) had MutL. 81 (2.4%) had other MMR IHC loss. IHC-PD-L1, TMB, neoantigen load (high affinity epitopes: p < 0.01, intermediate: p < 0.0001, low: p < 0.0001), TIS, IFN-gamma scores, immune related gene expressions, TME (Macrophage M1, CD8+) were higher in MutS. The mOS in MutS (N = 94) vs. MutL (N = 1804) was NR vs. 47 m (p < 0.001). In ICI-treated pts, the mOS in MutS (N = 11) vs. MutL (N = 273) was NR vs. NR (p = 0.559). Conclusions: This is the largest study to explore differential response to ICIs in CRC and EC pts with different MMR alterations. In pts with CRC and EC, the mOS was longer in MutS compared to MutL. In ICI-treated pts, the mOS was longer in MutS compared to MutL in CRC but not in EC. Among the explored biomarkers, neoantigen load was higher in MutS compared to MutL in both CRC and EC and maybe the driving factor for differential response to ICIs.

DEFB1 gene expression and the molecular landscape of colorectal cancer (CRC)

3523

Background: Defensins are antimicrobial peptides that play important roles in innate immune response. Deregulation of beta-defensin-1 ( DEFB1) gene expression has been implicated in several cancers and we previously showed that single nucleotide polymorphisms in DEFB1 are associated with clinical outcomes in patients with metastatic CRC. Hence, we aimed to further characterize the molecular features associated with DEFB1 gene expression in CRC. Methods: 14416 CRC tumors tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing on DNA (592 genes or WES), RNA (WTS) and IHC were analyzed. Top quartile transcripts per million (TPM) for DEFB1 expression were considered high (Q4) while bottom quartile low (Q1). Consensus molecular subtypes (CMS) were assessed using RNAseq. Cell infiltration in the tumor microenvironment (TME) was estimated by QuantiSEQ. X2/Fisher-Exact were used and significance was determined as P-value adjusted for multiple comparison ( Q <.05). Real-world overall survival information was obtained from insurance claims data and Kaplan-Meier estimates were calculated for molecularly defined patients. Results: DEFB1 expression was highest in left-sided and rectal tumors (median TPM 2.27) and lowest in right-sided tumors (median TPM 1.62). Overall, when compared to low expression, high DEFB1 was negatively associated with high TMB (≥ 10 Mut/Mb) (4.8% vs 17%), dMMR/MSI-H (2.2% vs 13.1%), and PD-L1 expression (3.1% vs 5.2%) (all Q <.05). Additionally, DEFB1 high was associated with higher expression of immune checkpoint genes CD274, CD80, CD86, HAVCR2, LAG3, PDCD1 and PDCD1LG2 (Fold Change/FC: 1.27-1.56) but lower IDO1 (FC: 0.89) (all Q <.05). Similar results were confirmed in MSS tumors only, but IDO1 was now positively associated with DEFB1 high (FC: 1.23). In the MSS cohort, DEFB1 expression was highest in CMS2 and lowest in CMS3 (2.84 vs 1.67 median TPM, Q <.05). In the MSS cohort, APC mutations were more frequent in DEFB1 high tumors (79% vs 72%) while BRAF (5.8% vs 9.4%), GNAS (1.1% vs 4.4%), FBXW7 (7.8% vs 10.5%), SMAD4 (12.3% vs 17%), RNF43 (2.2% vs 3.5%) and POLE (0.2% vs 0.7%) mutations as well as MYC (1.2% vs 2.6%) and MYB amplifications (0.1% vs 0.9%) were less frequent in DEFB1 high (all Q <.05). Higher neutrophils, NK cells, M2 macrophages, CD4+ T cells and myeloid dendritic cells but lower M1 macrophages, Tregs and CD8+ T cells in the TME were significantly associated with high DEFB1; both in the overall and MSS cohorts ( Q <.001). CRC patients with DEFB1 expression level above the median had worse OS compared to those below the median both in the overall cohort (HR: 1.18, 95% CI: 1.10-1.27) and in MSS tumors (HR: 1.18, 95% CI: 1.10-1.27). Conclusions: Our data show a distinct molecular landscape, including mutational profiles, CMS, immune biomarkers, and TME cell infiltration associated with DEF1B gene expression in CRC. These findings suggest a key role for DEF1B in modulating anti-tumor immunity and TME.

Landscape of endocytosis pathway in colorectal cancer (CRC)

3148

Background: Recent proteogenomic analyses of CRC revealed that driver gene alterations are enriched in the endocytosis pathway (Vasaikar S, et al. Cell 2019;177:1035-49). Endocytosis is a cellular system involving post-translational modification of plasma membrane proteins through internalization, intracellular trafficking, degradation, and recycling. Clathrin-mediated endocytosis (CME) is the main endocytic portal, and endosomal sorting complexes required for transport (ESCRT) play a critical role in the lysosomal degradation pathway. Besides the well-known function of endocytosis attenuating signaling pathways through receptor clearance from the cell surface, the opposite function contributing to signal maintenance has also been reported. However, the clinical implications of the endocytosis pathway alterations in CRC are largely unclear. Methods: We retrospectively reviewed CRC patient samples (n = 15025) submitted to a commercial CLIA-certified laboratory (Caris Life Sciences, Phoenix AZ). Next-generation sequencing of DNA and RNA (whole-transcriptome sequencing) and immunohistochemistry (IHC) were performed. CME-related (47 genes) and ESCRT-related (35 genes) expression signatures were calculated as composite z-scores and compared between subgroups stratified by RAS/ BRAF mutation status, MSS/MSI status, tumor sidedness, and consensus molecular subtype (CMS). VPS4A/ VPS4B expression correlation with major oncogenic pathway signatures (composite z-scores) and CMTM6/ CMTM4/ HIP1R expression association with PD-L1+ IHC were also assessed. Results: Among 17 endocytosis-related genes, no pathogenic/likely pathogenic mutations were identified. The CME-related signature was increased in RAS/ BRAF wild type vs. mutant (0.93 z-score difference, p= 0.04) and MSS vs. MSI-high (6.0 z-score difference, p< 0.01), while the ESCRT-related signature was higher in MSS compared to MSI-high (2.7 z-score difference; p< 0.01). No differences between tumor sidedness were observed in both CME- and ESCRT-related signatures (0.81 and 1.17 z-score differences, respectively). CMS4 had the highest expression of both signatures, while CMS3 had the lowest, of both CME- and ESCRT-related genes (each > 20 z-score difference, p< 0.01). VPS4A and VPS4B expression had a strong positive correlation with WNT, EGFR/MAPK, TGF-beta, and Notch pathway signatures (0.65-0.83 Spearman, all p< 0.01). CMTM6 expression was positively associated with PD-L1+ IHC (1.2-fold increase vs PD-L1-negative, p< 0.01), while CMTM4 and HIP1R expression showed a negative association (0.7- and 0.9-fold decrease, respectively, p< 0.01). Conclusions: This large study indicates endocytosis pathway expression is positively associated with oncogenic pathway signaling in CRC. Further analysis of RAS/ BRAF wild type, MSS, and CMS4 patient subgroups are warranted to determine the efficacy of targeting endocytosis pathways in CRC.

Characterization of TIM3 and its ligands in colorectal cancer

3547

Background: TIM-3 is an inhibitory checkpoint glycoprotein found on innate and adaptive immune cells and is highly expressed on tumor infiltrating lymphocytes. TIM-3 and its ligands, Galectin 9 (Gal9), HMGB1 and CEACAM1 play a critical role in immune regulation and preclinical data suggest a role in the pathogenesis of colorectal cancer (CRC). We aimed to characterize the molecular features and prognostic value of TIM3 and its ligands in CRC. Methods: Tumor molecular profiling was performed from 15,026 FFPE samples by NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) and immunohistochemistry (IHC) at Caris Life Sciences (Phoenix, AZ). Top quartile transcripts per million (TPMs) for TIM-3, Gal9, HMGB1 and CEACAM1 were considered high (Q4) while bottom quartile low (Q1) expression (exp). X2/Fisher-exact tests were used for comparison and significance was determined as P-value adjusted for multiple comparison and this was found for the results reported here ( Q < 0.05). Cell infiltration in the tumor microenvironment (TME) was estimated by quanTIseq. Real-world overall survival (OS) information was obtained from insurance claims data and Kaplan-Meier estimates were calculated for molecularly defined pts. Results: Gal 9/TIM3-high tumors had higher prevalence (prev) of high tumor mutational burden (TMB ≥ 10 Mut/Mb) (12% vs. 8%; 14% vs. 6%), deficiency in mismatch repair (dMMR) (9% vs. 5%; 10% vs. 4%), PD-L1 exp (5% vs. 3%; 7% vs. 2%), and was highest in transverse (Fold Change; FC: 1.05, 1.12) and right sided tumors (1.04, 1.10) compared to left sided tumors, and CMS4. In contrast, HMGB1 and CEACAM1-high tumors had lower prev of dMMR (5% vs. 8%, 3% vs. 12%), PD-L1 exp (3% vs. 5%, 3% vs. 6%) and TMB-H (8% vs. 11%, 6% vs. 16%), and was highest in CMS2. In MSS tumors, Gal9/TIM3-high tumors were associated (assoc) with lower frequency of TP53, amplifications (amp) of FLT1/3, CDX2, FOXO1, and CDK8 while CEACAM1 and HMGB1-high were assoc with higher mutation (mut) rates of TP53, APC, NRAS, amp of FLT1/3, CDX2, CDX8, and lower mut in GNAS, FBXW7 and RNF43. High Gal9 and TIM3 exp was assoc with higher infiltration of B cells, M1 and M2 macrophages, NK cells, CD4+ and C8+ T cells, and Tregs, while high HMGB1 and CEACAM1 exp was negatively assoc with Tregs, M1 macrophages, monocytes and CD8+ T cells. High exp of Gal9, TIM3, HMGB1, and CEACAM1 was assoc with worse OS in the entire cohort (HR 0.90, 95%CI, 0.84-0.97, P = 0.005, HR 0.81, 95% CI, 0.75-0.87, P< 0.00001; HR 0.88, 95% CI, 0.82-0.95, P < 0.001; HR 0.80; 95%CI, 0.74-0.86, P < 0.00003, respectively). Conclusions: Strong assoc were identified between Gal9/TIM3, HMGB1 and CEACAM1 gene exp and IO biomarkers, distinct molecular features, CMS, TME cell infiltration, and patient outcomes in CRC. Significantly different mut frequencies may signify unique subsets of CRC. These findings provide rationale for further evaluation of TIM3 and its ligands in CRC as prognostic biomarkers and potential therapeutic targets modifying the TME.

The prognostic significance of TP53 mutations in patients with right-sided and left-sided colorectal cancer

3589

Background: In patients with colorectal cancers (CRCs), prior studies have reported that various TP53 mutations have prognostic significance. The anatomic location of the primary CRC and the TP53 mutation subtype influence patient survival. In this study, we explored the prognostic significance of TP53 mutations (mTP53) classified as gain-of-function (GOF) or non-GOF in patients with right-sided (RCC) and left-sided CRCs (LCC). Methods: CRC specimens (6,248 RCCs and 14,215 LCCs) were tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing (NGS) of DNA (592-gene panel or whole- exome sequencing). R175H, R248W, R248Q, R249S, R273H, R273L, and R282W were defined as GOF mTP53 and all other mTP53 were defined as non-GOF mTP53. MSI-H/dMMR status was determined by immunohistochemistry (IHC) of MMR proteins and/or NGS. Real-world median overall survival (mOS) was obtained from insurance claims data and calculated from tissue collection to last contact using Kaplan-Meier estimates. Results: GOF mTP53 and non-GOF mTP53 were identified in 15% and 39% respectively, in RCC and 17% and 46% respectively, in LCC. In RCC, the mOS for patients with wild-type TP53 ( wtTP53) vs. GOF mTP53 was 34 months (m) vs. 23m (p < 0.00001), and the mOS for patients with wtTP53 vs. non-GOF mTP53 was 34m vs. 27m (p < 0.001). In LCC, the mOS for patients with wtTP53 vs. GOF mTP53 was 35m vs. 33m (p = 0.056), and the mOS for patients with wtTP53 vs. non-GOF mTP53 was 35m vs. 35m (p = 0.32). The mOS for patients with non-GOF mTP53 vs. GOF mTP53 in RCC and LCC was 28m vs. 24m (p = 0.096), and 35m vs. 34m (p = 0.175), respectively. The prognostic value of GOF mTP53 and non-GOF mTP53 was further explored in relation to MSI-H/dMMR, RAS, BRAF, and PIK3CA mutation status. The worse prognosis associated with mTP53 in RCC was seen in all comparisons, except in GOF mTP53/MSI-H/dMMR, and non-GOF mTP53/ wtKRAS subgroups. In patients with LCC, worse prognosis associated with GOF mTP53 and non-GOF mTP53 was only noticeable in KRAS and PIK3CA mutant subgroups. Conclusions: This is the largest study to explore TP53 mutations and their prognostic significance in patients with RCC and LCC. The prevalence of GOF mTP53 and non-GOF mTP53 was higher in LCC compared to RCC. However, both GOF mTP53 and non-GOF mTP53 were associated with worse mOS for patients with RCC, but not LCC. Our study validates the sidedness-dependent prognostic significance of TP53 mutations. It also shows that the worse prognosis of mTP53 is independent of the approach of collectively classifying TP53 mutations into GOF vs non-GOF. Given the sheer extent and diversity of TP53 mutations, a more nuanced approach towards re-classification of GOF mTP53 is warranted. Detailed information on p53 mutations will be crucial for the interpretation of future clinical trials and for the design of novel therapeutic strategies.

HER2 alterations and prognostic implications in all subtypes of breast cancer

1041

Background: Amplification or overexpression of human epidermal growth factor receptor 2 (HER2) oncogene is present in about 15-20% of breast cancers & is a prognostic & predictive biomarker. Additional ERBB2/HER2 alterations have become apparent on tumor next generation sequencing (NGS), including activating kinase domain mutations & fusions. Methods: DNA NGS (592 gene panel or whole exome) data from 12,153 breast samples retrospectively reviewed for ERBB2 alterations with RNA whole-transcriptome sequencing (WTS) data available for 7289 (60%) samples. Gene fusions detected using the ArcherDx fusion assay or WTS. Clinicopathologic features were described including breast cancer subtype, age, & biopsy site. HER2 status determined according to 2018 ASCO-CAP guideline. Overall survival obtained from insurance claims & Kaplan-Meier estimates were calculated for defined patient (pt) cohorts. Statistical significance was determined using Chi-square & Wilcoxon rank sum tests. Results: ERBB2 mutations ( ERBB2mts) were identified in 3.2% (n = 388) of tumors overall & most common in liver metastases (113/1972, 5.7%). ERBB2mts were found more in breast lobular tumors compared to ductal tumors (10 vs 2.1%, p < 0.001). HER2+ tumors had higher frequency of ERBB2mts compared to HER2- (4.3 vs 3%, p = 0.028). Tumors with score of 0 by immunohistochemistry demonstrated lower rate of ERBB2mts (0+ 2.2%, 1+ 3.5%, 2+ 4.5%, 3+ 3.45%, p < 0.05). Among HER2- tumors, ERBB2mts were present in 3.6% of hormone receptor (HR)+/HER2- & 1.9% of TNBC. Metastatic tumors had a higher rate of ERBB2mts compared to locoregional breast tumors (3.8 vs 2%, p < 0.001), with increased rates of activating mutations S310F (0.1 vs 0.0%, p < 0.05) & D769H (0.3 vs 0.1%, p < 0.05), & the resistance mutation L755S (1.2 vs 0.6%, p < 0.01). Compared to ERBB2-WT, ERRB2mts were associated with decreased ERBB2 transcripts levels in HER2+ samples (222 vs 441 transcripts per million [TPM], p < 0.001) & increased levels in HER2- samples (73 vs 35 TPM, p < 0.001). High tumor mutational burden (≥ 10 mut/Mb) & ERBB3 mutations were more common in ERBB2mts compared to ERRB2-WT (16.7 vs 7.7%, p < 0.001; 10.6 vs 0.8%, p < 0.001). ERBB2 fusions were rare (0.49%) with 97% occurring in HER2+ tumors. Of 8358 pts with outcome data, prognosis (HR 1.2, P = 0.06) & response to chemotherapy (HR 1.1, P = 0.42) was similar between pts with HER2- ERBB2mt & ERBB2-WT. Conclusions: ERBB2mts & fusions were observed in all breast cancer subtypes - more commonly in HER2+, metastatic, & lobular histology tumors - & did not influence prognosis. These alterations may reflect response to treatment pressures in HER2+ disease to reactivate HER2-mediated growth pathways following anti-HER2 therapy & may represent a targetable upregulated oncogenic pathway in HER2- disease. Ongoing identification of ERBB2 alterations may augment treatment options for breast cancer pts & clinical outcomes from this approach are under investigation.

Surfaceome profiling to reveal unique therapeutic vulnerabilities in transcriptional subtypes of small cell lung cancer (SCLC)

8515

Background: Effective treatment options for SCLC remain limited and new treatment approaches are needed to improve outcome. We sought to validate the initial observation in cell lines and limited tissue samples of SCLC of a differential expression of cancer/testis (CT) antigens and TACSD2 gene that encodes surface protein, Trop2 across various subtypes of SCLC. We also tested whether overall surfaceome profile as previously described in other tumor types will show hierarchical priority of expression between transcriptionally defined SCLC subtypes. Methods: We conducted a comprehensive surfaceome profiling of SCLC samples using data generated by RNA sequencing (whole transcriptome) at Caris Life Sciences (Phoenix, AZ). SCLC tumors were stratified into 5 subgroups (SCLC-A/N/Y/P and -mixed) based on the relative expression of the four transcription factors. Expression values were converted to z-scores (the expression value for each gene is normalized to the average expression of that specific gene such that the z-score reflects the number of standard deviations above or below the average). The highest positive z-score among the 4 transcription factors determined subgroup. If all transcription factor z-scores for a given sample were negative, the sample was assigned to ‘Mixed’ subgroup. Significance was tested by Chi-square, Fisher’s exact test, or Mann-Whitney U test. Results: We employed data generated from 674 SCLC samples; median age of 66 years and male (48.7%). The SCLC subtype distribution was 241 (35.8%), 120 (17.8%), 40 (5.9%), 143 (21.2%), 130 (19.3%) for types A, N, P, Y and mixed respectively. Supervised analysis for TACSTD2 expression showed highest levels in YAP1 subtype and was overall significantly increased in SCLC-Y (̃3-fold) and SCLC-P (̃2-fold) subtypes compared to A, N and mixed subtypes. Similarly, SCLC-Y subtype showed the highest median expression as well as the strongest correlation with most TACSTD2-interacting and regulatory genes. A top 10 list of candidate surface protein gene out of 3699 surfaceome genes was defined for each subtype based on the strength of correlation. The top candidate surface protein gene and CT antigen gene respectively by subtype were: SCN3A (r = 0.7033, p = 1.08E-101) and NOL4, (r = 0.574, p = 2.46E-60) for SCLC-A; SSTR2, (r = 0.742, p = 8.18E-119) and TMEFF1, (r = 0.3601, p = 4.53E-22) for SCLC-N; TMPRSS13 (r = 0.5699, p = 2.64E-59) and LY6K (r = 0.4778, p = 9.80E-40) for SCLC-P; and CYBRD1 (r = 0.8559, p = 1.18E-194) and CTAGE5 (r = 0.5521, p = 4.95E-55) for SCLC-Y. Conclusions: SCLC-Y subtype showed the highest expression of TACSTD2 and its interacting and regulatory genes. This subtype could serve as an enrichment factor for antibody-drug-construct targeting TROP2. Several candidate CT antigens and surfaceome genes showing strong correlation with lineage-defining transcription factors offer additional therapeutic targets in SCLC.

Characterization of NY-ESO-1 gene expression in gastric cancer (GC)

4046

Background: Tumor specific antigens, such as NY-ESO-1 , are emerging as key tumor immune modulating factors with great potential to be used as therapeutic targets to enhance immunotherapy efficacy and expand treatment options for GC. We sought to compare GC tumors expressing high vs low levels of NY-ESO-1 in terms of immune cell abundance in the tumor microenvironment (TME) as well as distinct molecular features and immune biomarkers. Methods: 1967 tumor samples tested with NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) by Caris Life Sciences (Phoenix, AZ) were retrospectively reviewed. The top quartile of transcripts per million was considered high while the bottom quartile was considered low NY-ISO-1 expression. Tumor mutational burden (TMB) was calculated based on somatic nonsynonymous mutations. Mismatch repair deficiency/microsatellite instability (dMMR/MSI) status was evaluated by a combination of IHC, fragment analysis and NGS of known MSI loci. Gene fusions were detected based on WTS. X2, Fisher-exact, and Mann Whitney tests were used for comparison and significance adjusted for multiple testing by Benjamini-Hochberg correction ( q < 0.05). Cell infiltration in the TME was estimated by quanTIseq. Gene expression profiles were analyzed for a transcriptional signature predictive of response to immunotherapy (T cell-inflamed signature: TIS). Results: The analysis was focused on primary tumors (N = 1323) in this initial study. Expression of NY-ESO-1 was lower in primary/local than metastases (Fold Change FC met vs primary: 1.60, q < 0.05). NY-ESO-1 expression did not appear to be strongly associated with distinct gene mutation profiles in GC. There were no significant differences between low and high expression of NY-ESO-1 with regards to well established immuno-oncology markers (dMMR/MSI, TMB, PD-L1). However, high NY-ESO-1 expression was positively associated with immune related gene expression including CD274, CD80, CD86, CTLA4, HAVCR2, IDO1, IFNG, LAG3, PDCD1, and PDCD1LG2 (FC low vs high: 0.56 to 0.79, q < 0.0001). High NY-ESO-1 expression was also positively associated with cell abundance in the TME including NK cells (FC = 0.87, q < 0.0001), monocytes (FC = 0.29, q < 0.05), myeloid dendritic cells (FC = 0.66, q < 0.0001), CD4+ non-reg T cells (FC = 0.54, q < 0.0001), and CD8+ T cells (FC = 0.73, q < 0.05). Similarly, tumors with high NY-ESO-1 expression were associated with higher TIS scores ( q < 0.0001). Conclusions: In our large cohort of GC, tumors expressing high NY-ESO-1 displayed a distinct landscape of immune cells in the TME and were associated with high expression of immune related genes, as well as high TIS score, which has been reported to predict benefit from anti-PD-1 treatment. The results of our analysis support an association between a more immunologically active tumor microenvironment and NY-ESO-1 gene expression which may have relevant implications on immunotherapy treatment for GC.

Genomic analysis of clear cell carcinoma

5548

Background: Clear cell carcinomas (CCC) are rare histologies outside of the kidney and are typically less sensitive to standard treatments. Genomic alterations in chromatin remodeling pathways involving ARID1A or the intracellular PI3K-mTOR signaling pathway are found in both renal and ovarian CCC. It is unclear whether CCCs originating from different anatomic sites share a common genomic landscape. This CARIS Precision Oncology Alliance project sought to determine whether CCC of different organs shared similar genomic signatures and to identify potential pathways that could be targeted in a tumor-agnostic clinical trial. Methods: CCCs (N = 861) from multiple primary tumor sites, including kidney (30.5%), ovary (39%), endometrium (23.9%), other gynecologic sites (e.g., cervix, fallopian tube, 3.3%), and miscellaneous (non-kidney or gynecologic sites, 3.3%) were analyzed at the Caris Life Sciences Laboratory (Phoenix, AZ). Using hierarchical clustering (HC) and principal component analysis (PCA), the samples were compared across 648 total genes from five metabolic gene sets consisting of angiogenesis, glycolysis, hypoxia, oxidative phosphorylation, and fatty acid metabolism. Gene Set Enrichment Analysis (GSEA) was further conducted on the samples across fifty hallmark gene sets representing specific biologic processes and expression. Samples were also analyzed for individual genomic alterations and immune-oncology associated biomarkers. PD-L1 (SP142) expression was evaluated by immunohistochemistry (positive threshold: 2+ stain intensity and ≥ 5% tumor cells). Results: HC and PCA demonstrated that renal CCC formed distinct clusters compared to non-renal CCC. Tumors from gynecologic sites could not be separated into distinct clusters. GSEA showed that the hypoxia gene set was significantly upregulated in the renal but not in non-renal CCCs. Mutations involving TP53, ARID1A, PIK3CA were found to be the most altered genes in endometrial (62%, 26%, 31%), ovarian (13%, 55%, 48%), other gynecological sites (33%, 38%, 44%), and non-gynecologic CCC (13%, 17%, 12%) respectively. PD-L1 expression, high tumor mutational burden (≥10 mutations/Mb), and deficient mismatch repair/microsatellite instability rates across sites were: kidney (11%, 2%, 2%), endometrium (13%, 12%, 7%), ovary (9%, 4%, 3%), other gynecological sites (31%, 11%, 11%), and miscellaneous sites (11%, 19%, 4%). Conclusions: Initial metabolic gene expression clustering analysis shows that CCCs do not separate by organ of origin beyond renal versus extra-renal. TP53, ARID1A, and PIK3CA were the most frequently altered genes in non-renal CCC. Out of fifty hallmark gene sets, only two were statistically significantly different among gynecological CCCs. This similarity between gynecological CCC can be leveraged by targeting pathways such as PI3K-AKT-mTOR, DNA repair, and MYC targets in a site agnostic manner. Furthermore, high PD-L1 expression is found in other gynecological sites.

Molecular correlates of MAEA expression in colorectal cancer (CRC)

3128

Background: Macrophage Erythroblast Attacher (MAEA) plays an important role in actin cytoskeleton rearrangement in macrophages and erythroid cells. We previously reported that MAEA suppresses migration, invasion and enhances chemosensitivity in CRC cell lines. Here we aimed to characterize the molecular features associated with MAEA gene expression in CRC. Methods: 14416 CRC were tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS). Top quartile transcripts per million (TPM) for MAEA expression were considered high (Q4) while bottom quartile low (Q1). Consensus molecular subtypes (CMS) were assessed using RNAseq. Cell infiltration (CI) in the tumor microenvironment (TME) was estimated by QuantiSEQ. X2 and Fisher-Exact tests were used and significance was determined as P-value adjusted for multiple comparisons ( Q < 0.05). Results: MAEA expression was highest in rectal tumors (13.6 median TPM) followed by transverse and right-sided tumors (13.0 and 12.8, respectively) and lowest in left-sided tumors (12.5). Overall, MAEA TPM were associated with higher tumor mutational burden (≥ 10 Mut/Mb) (11.8% vs. 8.2%) and dMMR/MSI-H (8.7% vs. 5.1%) ( Q < 0.0001); however, the association with TMB was not observed in MSS tumors. In the MSS cohort, MAEA expression was the highest in CMS4 (14.9 median TPM) followed by CMS1 (12.5), CMS2 (11.9), and the lowest in CMS3 (10.3, all intergroup Q < 0.05). MAEA high was associated with lower mutation rates of APC and amplification of FLT1/ FLT3 while higher mutation rates of ASXL1, KMT2A/C/D, SMARCA4, FBXW7, PTEN, RNF43, BRCA2, HNF1A in the overall cohort ( Q < 0.05). In the MSS cohort, FBXW7 mutation significance with MAEA high expression held true ( Q < 0.05) while MAEA high expression trended to associate with higher mutation rates of KMT2D, SMARCA4, PTEN, BRCA2 mutations, and a lower frequency of FLT1/ FLT3 CNA ( P < 0.05 but Q > 0.05). High MAEA was associated with higher immune CI in the TME, including B cells, macrophages (M1 and M2), neutrophils, NK cells, Tregs, CD4+ T cells and myeloid dendritic cells both in the overall cohort and in MSS tumors (fold change: 1.11-1.33, all Q < 0.001). Conclusions: Our data show a strong association between MAEA gene expression and distinct molecular features (including CMS and immune biomarkers) and TME cell infiltration in CRC. These findings suggest that targeting MAEA may have relevant clinical applications in selected CRC subgroups and MAEA may be an important player in determining the composition of the TME.

Comprehensive profiling of clock genes expression in colorectal cancer (CRC)

3129

Background: Disruption of the circadian clock has been linked to cancer risk, development and progression. Core clock proteins are emerging as novel therapeutic targets in cancer. We previously showed that polymorphisms in clock genes were associated with anti-VEGF treatment outcome in metastatic CRC. Here we further evaluated the molecular landscape of clock pathway alterations in CRC. Methods: 7591 CRC tested at Caris Life Sciences (Phoenix, AZ) with WTS (Illumina NovaSeq) and NextGen DNA sequencing (NextSeq, 592 Genes and NovaSEQ, WES) were analyzed. Clock gene Score (CS) was determined using expression of core clock genes Z scores (positives of CLOCK, ARNTL, RORA/B/C and negatives of repressors CRY1/2, PER1/2/3, REVERBA/B) stratified by quartiles. xCell was used to quantify cell infiltration in the tumor microenvironment (TME). Consensus molecular subtypes (CMS) were assessed by RNAseq. Significance was determined as P-values adjusted for multiple testing ( q) of <.05. Real world survival was obtained from insurance claims data and Kaplan-Meier estimates were calculated for comparison. Results: CS was higher in primary tumors than metastases and in right- than left-sided CRC ( P <.001). Liver metastases were associated with lower CS (23% Q1 vs 19% Q4, P <.001). CS was positively associated with CMS1 and 3 (21 vs 11% and 23 vs 9%, respectively, Q4 vs Q1) and negatively correlated with CMS2 and 4 (22 vs 32% and 34 vs 48%) (all P <.001). These associations were confirmed in mismatch repair proficient (pMMR) tumors. Overall, TMB-H and dMMR/MSI-H were positively associated with CS (11 vs 6% and 8 vs 4%, Q4 vs Q1, q <.0001) and PD-L1 showed a similar trend ( P <.01, q =.06); the association with TMB-H was not significant in pMMR. High CS was associated with alterations of genes in WNT signaling, RAS, PI3K, TGF-β, and NOTCH pathways, while negatively associated with TP53 mutations, HER2 expression and CDX2 copy numbers, confirmed in pMMR (all q <.05). CS negatively correlated with the angiogenesis pathway signature (Q1 vs Q4 Z score: 6.6 vs -4.6, P <.001). B cells, M1 and M2 macrophages, neutrophils, NK, Tregs, CD4+ and CD8+ T cells, and myeloid dendritic cells were more abundant in the TME of tumors with high CS while cancer associated fibroblasts were lower, regardless of MMR status (all q <.001). Individually, ARNTL tumor expression below median was associated with better OS (overall: HR 0.88, 95% CI [0.82-0.94]; pMMR: HR 0.88 [0.81-0.94]) and longer time on treatment of bevacizumab (overall: HR 0.91 [0.83-0.99]; pMMR: HR 0.91 [0.83-0.99]). Conclusions: This is the most extensive profiling study to investigate the expression of clock genes in CRC. Our data show that clock genes expression is strongly associated with distinct molecular features, immune cell infiltration, angiogenesis pathway enrichment and patient outcomes. These findings support the clock pathway as a therapeutic target in CRC, with a major role in CRC biology and TME modulation.

Claudin 18 (CLDN18) gene expression and related molecular profile in gastric cancer (GC)

4048

Background: Claudins are transmembrane proteins which maintain the tight junction between cells. The stomach specific isoform, CLDN18 isoform 2 (CLDN18.2), is emerging as a promising treatment target because of high expression in GC cells, including targeting via adoptive T-cell strategies. We characterized the molecular features associated with CLDN18 isoform 1 and 2 ( CLDN18.1/ 18.2) gene expression in GC. Methods: Tumor profiling was performed from 1967 samples by NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) at Caris Life Sciences (Phoenix, AZ). EBER (Epstein Barr Virus) was tested by CISH. Top quartile transcripts per million for CLDN18.1/18.2 were considered high while bottom quartile low expression. X2, Fisher-exact, and Mann Whitney tests were used and significance adjusted for multiple testing by Benjamini-Hochberg (q <.05). Cell infiltration in the tumor microenvironment (TME) was estimated by quanTIseq. Gene expression profiles were analyzed for a transcriptional signature predictive of response to immunotherapy (T cell-inflamed signature, TIS). Results: CLDN18.2 expression was detected in 97% of the samples and CLDN18.1 in 63%. Primary tumors had significantly higher expression levels of both CLDN18.1/18.2 (Fold Change 18.1: 0.50; 18.2: 0.65), compared to metastatic tumors ( p <.05), thus we focused on the comparison of CLDN18 high and low in primary tumors. CLDN18.2 high expression group had higher CLDN18: ARHGAP26 fusion positive rate (low vs high: 0.91% vs 5.5%, q <.0001), and a trending association with CDH1 mutation (11.7% vs 20.7%, p <.01, q >.05) and EBER (2.15% vs 6.31%, p <.05, q >.05). There were more prevalent ARHGAP26 fusions in the CLDN18.1/18.2 high group (18.1: 9.5% vs 3.86%, p =.001; 18.2: 10.1% vs 0.9%, q <.0001), with the most common fusion between CLDN18 exon 5 and ARHGAP26 exon 12. CLDN18.2 high expression demonstrated an inverse trending correlation with PD-L1 (24.9% vs 18.3%, p <.05; q >.05) and TMB-H (19.6% vs 12.2%, p <.05; q >.05). Similarly, CLDN18.1/18.2 displayed an inverse relationship with M1 Macrophages, NK cells, CD4+ T cells, myeloid dendritic cells in the TME ( q <.05); with higher CLDN18 expression associated with fewer immune cells and a colder TME, especially in isoform 2. The TIS score was significantly higher in the CLDN18.2 high expression group ( q <.05), but lower in CLDN18.1 high expression group ( q <.0001), respectively. Conclusions: This is one of the most comprehensive dedicated analyses on CLDN18 related to tumor molecular features, TME and immunotherapy response in GC. Tumors expressing high CLDN18, especially 18.2, displayed distinct genomic and transcriptomic alterations in immune biomarkers and immune cell infiltration in the TME. Anti-CLDN18.2 monoclonal antibodies are being tested in GC and CLDN18 is a target for ADC and CAR-T therapies. Our data suggest that expression may play a role in guiding patient selection and treatment combinations.

Age as a factor in the molecular landscape and the tumor-microenvironmental signature of osteosarcoma

11525

Background: Osteosarcoma (OS) incidence is characterized by a bimodal age distribution, with peaks in early adolescence and in adults > 65 years of age. In contrast to adolescents, OS in adults is frequently considered as a secondary neoplasm (i.e., transformation of Paget´s disease of the bone, radiation induced). Yet, the literature is scarce regarding the impact of age on the molecular landscape of OS. Herein, we sought to explore the association between age and the genomic profile as well as the tumor immune microenvironment (TME) in a large cohort of OS patients. Methods: 208 specimens were centrally analysed at the Caris Life Sciences laboratory with DNA seq (NextSeq, 592 gene panel or NovaSeq, whole-exome sequencing), RNA seq (Archer fusion panel or whole-transcriptome sequencing) and immunohistochemistry (IHC). RNA deconvolution and differential expression analyses were performed using the Microenvironment Cell Populations counter method for quantification of immune cell populations and gene expression profiling. The cohort was stratified into three distinct age groups (< 25 years [n = 83], 25-45 years [n = 58], > 45 years [67]). Results: Overall, the most frequently detected mutations were in TP53 (37%), RB1 (13%), ATRX (9%), TERT (6%), PTEN (5%), PIK3CA (4%) and KMT2D (3%). Copy number alterations were most frequently detected in CDK4 (12%), LRIG3 (11%), FLCN (11%), MDM2 (9%), CCND3 (9%), VEGFA (8%), TFEB (8%). Interestingly, age-based stratification revealed an increased frequency of FLCN (19.7 vs 4.7%, p < 0.01), CCND3 (13.9 vs 3.1%, p < 0.05), and HSP90AB1 (11.3 vs 0.0%, p < 0.01), alterations in patients < 25 years compared to > 45 years. TME analysis revealed that patients > 45 years have decreased B-cell abundance compared to patients < 25 years (2.9-fold decrease, p < 0.05) and 25-45 years (4.8-fold decrease, p < 0.05). Although not statistically significant, median transcriptional expression of PD-L1 was numerically increased in patients > 45 years (1.8-fold compared to 25-45 years, p = 0.17; 2.0-fold compared to < 25 years, p = 0.27), which was consistent with increasing rates of IHC PD-L1 expression with age (5.3%, 9.4%, and 17.5%, respectively, p = 0.06). Conclusions: To the best of our knowledge, this study represents the largest cohort of molecularly characterized OS. Age-associated differences in the genetic landscape and TME composition, including increased gene amplifications observed in younger patients and decreased B-cell abundance in older patients, might suggest fundamental underlying molecular and biological differences.

Comprehensive characterization of PTPRT expression in colorectal cancer (CRC)

3538

Background: PTPRT is a protein coding gene involved in signal transduction and cellular adhesion. It acts as a tumor suppressor gene and mutated PTPRT has been implicated in the early metastasis of CRC. PTPRT mutations have been reported as independent potential biomarkers for bevacizumab resistance in metastatic CRC and linked to improved response and survival of patients treated with checkpoint inhibitors in several tumors. Here we characterized the molecular features and clinical outcomes associated with PTPRT gene expression in CRC. Methods: 15025 CRC tested with NextGen Sequencing on DNA (592 genes or WES) and RNA (WTS) by Caris Life Sciences (Phoenix, AZ) were analyzed. Top quartile transcripts per million for PTPRT expression were considered high (H) and bottom quartile low (L). Cell infiltration in the tumor microenvironment (TME) was estimated by QuantiSEQ. X2, Fisher-Exact, Mann-Whitney tests were used and significance determined as P-value adjusted for multiple comparisons ( Q <.05). Real world survival was obtained from insurance claims data and Kaplan-Meier estimates were calculated for molecularly defined patients. Results: PTPRT expression was higher in left- than right-sided CRC and in metastases than primary/local tumors; it was highest in CMS4 and lowest in CMS1 (all Q <.05). PTPRT mutants had lower expression than wild type (0.037 vs 0.064, Q <.05). Overall, PTPRT-H had a lower rate of TMB-H (5% vs 17%), deficient mismatch repair (dMMR) (3% vs 13%) and PD-L1 (2% vs 6%) (H vs L, all Q <.05). In the proficient MMR (pMMR) cohort, PTPRT-H remained inversely correlated with TMB and PD-L1, was negatively associated with rates of mutated KRAS, PIK3CA, SMAD4, FBXW7, and positively associated with mutated TP53 and CDX2 amplification (all Q <.05). Expression of immune related genes was higher in PTPRT-H CRC, including CD274, CD80, CD86, CTLA4, HAVCR2, IDO1, IFNG, LAG3, PDCD1, and PDCD1LG2, regardless of MMR status (all Q <.05). PTPRT-H was associated with higher immune cell infiltration in the TME including B cells, M2 macrophages, neutrophils, NK, Tregs, CD4+ and CD8+ T cells, and myeloid dendritic cells (fold change/FC: 1.21-7.1), but lower M1 macrophages (FC: 0.76), regardless of MMR status (all Q <.05) with the only exception of CD8+ T cells in dMMR. PTPRT expression above median was associated with better OS (overall: HR 0.69, 95% CI [0.64-0.74]; pMMR: HR 0.67 [0.63-0.73]), longer time on treatment of bevacizumab (overall: HR 0.80 [0.74-0.87], pMMR: HR 0.80 [0.74-0.87]), and shorter time on immunotherapy treatment in the dMMR cohort (HR 2.13 [1.33-3.45]). Conclusions: Our data show a strong association between PTPRT expression and distinct molecular features (including CMS and immune biomarkers), TME cell infiltration and targeted treatment outcomes in CRC. These findings support PTPRT as a candidate prognostic and predictive biomarker for bevacizumab and immunotherapy treatment, and as a potential target in CRC.

Capicua (CIC) mutations in gliomas in association with MAPK activation for exposing a potential therapeutic target

2056

Background: Capicua (CIC) Gene is a tumor suppressor, transcriptional repressor, and a member of the high mobility (HMG)-box protein family. CIC is a negative regulator of MAPK and RTK pathways; inactivating CIC mutations (mut) occur in approximately 40% of oligodendrogliomas (OLIG) and less frequently in other gliomas putatively activating downstream signaling. With a goal to identify potential novel treatment options for various gliomas, we explored key signaling pathways associated with CIC mut. Methods: Consecutive glioma tumors were analyzed using Next-Gen DNA sequencing (NextSeq, 592 genes or NovaSeq, whole-exome), RNA sequencing (NovaSeq, WTS ) and IHC (Caris Life Sciences, Phoenix, AZ). Immune cell fraction was calculated by QuantiSeq; MAPK activation score (MPAS) was evaluated using RNAseq data. A comparison was made using Chi2 or Fisher’s-exact test with correction for multiple-comparison (q) using Benjamini-Hochberg. Results: A total of 196 (3.7%) tumors with CIC mut were seen in 5266 gliomas tumors analyzed, with the highest prevalence seen in OLIG (143 of 285, 50.2%). There was no difference between grade 3 (73 of 142, 51.4%) and grade 2 OLIG (70 of 143, 49%). CIC mut were infrequent in astrocytomas (16 of 829, 1.9%; grade 3, 12/510 or 2.4%; grade 2, 4/261 or 1.5%; grade 1, 0/58). CIC mut were present in glioblastomas (24/2753 or 0.6%), gliosarcomas (1/128 or 0.8%), and other mixed subtypes (12/185 or 6.5%). CIC mut were associated with higher prevalence of IDH1/2 mut (92% in CIC-mut vs. 17% in wild type), MGMT promoter methylation (97% vs. 47%), FUBP1 mut (32% vs. 1%) but lower PTEN mut (1% vs. 25%) and TP53 mut (12% vs. 39%) (all q < 0.05). Significant mutual exclusivity for CIC mut and MAPK pathway drivers observed: EGFR amplification (1.5% vs. 27%) , EGFR mut (0.5% vs. 12.6%), NF1 mut (4% vs. 18%) (all q < 0.05). BRAF mut rate was similar in CIC-mut or wild-type (1% vs. 3%, p = ns). Although associated with a higher tumor mutation burden (cutoff > = 10 mut/MB, 13% vs. 3%), a lower prevalence of PDL1 expression (1% vs. 16%) and lower M1 macrophage infiltration were seen (all q < 0.05). Similar effects were seen when stratifying by oligodendroglial and astrocytic histology. CIC mut were associated with increased MPAS score in OLIG (p = 0.01), particularly when compared to tumors lacking additional MAPK drivers (p = 0.001). This effect was not seen in astrocytic tumors, although EGFR alterations (including CNA, EGFRvIII, EGFR fusion and mut) were independently associated with increased MPAS scores (p < 0.001). Conclusions: CIC mut were frequent in oligodendrogliomas but occurred rarely in other glioma tumors and are associated with favorable prognostic markers. RNA expression analysis suggests CIC mut is associated with MAPK activation in OLIG, as are EGFR alterations in astrocytomas. Targeted inhibition of this pathway in selected gliomas may be a promising therapeutic avenue and should be explored further.

CXCR4 overexpression: An indicator of poor survival and predictor of response to immunotherapy in patients with metastatic colorectal cancer

3546

Background: CXC-chemokine receptor 4 (CXCR4) is a ubiquitous chemokine receptor activated by the CXCL12 ligand and is implicated in tumor invasion, metastasis, and immune cell (IC) trafficking. High CXCR4 expression is associated with poor prognosis in colorectal cancer (CRC). < 10% of metastatic CRC cases harbor microsatellite instability (MSI-H) and demonstrate lower tumor mutation burden (TMB), decreased IC infiltration, and lack of response to current immunotherapy regimens. This study aims to interrogate the role of CXCR4 mRNA expression on the the tumor microenvironment (TME) and its prognostic and predictive value to tailor immunotherapeutic treatment strategies in CRC. Methods: A total of 15,026 CRC samples were analyzed using whole-exome sequencing, whole-transcriptome sequencing, and immunohistochemistry (Caris Life Sciences, Phoenix, AZ). Study cohort was stratified by CXCR4 mRNA expression levels in quartiles (Q1 (low) vs Q4 (high)). IC fraction was calculated by QuantiSeq, and real-world overall survival information was obtained from insurance claims data and calculated from tissue collection time to last day of contact. Statistical significance was determined using chi-square/Fisher-Exact and adjusted for multiple comparisons (q < 0.05). Results: Samples obtained from metastatic sites showed higher CXCR4 mRNA expression than those from primary tumors (22.7 vs 18.6 median transcripts per million (TPM), p < 0.001). CXCR4 mRNA expression was significantly lower in liver metastases than in non-liver metastases (21.2 vs 24.8 TPM, p < 0.001). Median CXCR4 mRNA expression was highest in the consensus molecular subtypes 4 (33.3 TPM) and lowest in 3 (13.0 TPM, p < 0.05). CXCR4 mRNA expression was positively associated with TMB-H, MSI-H/dMMR, and positive PD-L1 IHC status. In the TME, high CXCR4 mRNA expression was observed in tumors with a higher IC infiltration including B cells, M1/M2 macrophages, NK cells, CD8+ T cells and T-regs, regardless of MSI status. High CXCR4 mRNA expression in the primary tumor was associated with poor prognosis (HR 0.77, 95% CI 0.70-0.85; p < 0.001), regardless of MSI-status. In metastatic tumors, low mRNA expression was correlated with improved survival (HR 0.89, 95% CI 0.80-0.99; p = 0.34); however, this did not reach statistical significance in the MSS cohort (HR 0.90, 95% CI 0.80-1.0; p = 0.06). Of note, high CXCR4 mRNA expression was associated with improved survival in all patients with CRC who received pembrolizumab (HR 2.12, 95% CI 1.16-3.91; p = 0.013). Conclusions: This is the largest clinical dataset to date demonstrating high CXCR4 expression as a predictor for poor survival in CRC. Furthermore, high CXCR4 expression was associated with improved outcome after checkpoint inhibition immunotherapy, indicating its strong potential as a predictive biomarker that could inform immunotherapeutic strategies in CRC.

Association of PAK4 expression with overall survival in patients with non-small cell lung cancer (NSCLC)

e21149

Background: P21-activated kinase 4 (PAK4) is a crucial effector of the Rho GTPases. It acts as a regulatory switch that controls a wide range of cellular functions and plays a pivotal role in cancer progression and metastasis. Very little is known about the expression and prognostic value of PAK4 in NSCLC. Methods: 17,689 NSCLC tumor samples were analyzed using next-generation sequencing (NextSeq, 592 Genes, or WES NovaSeq), immunohistochemistry (IHC), and whole transcriptome sequencing (NovaSeq) (Caris Life Sciences, Phoenix, AZ). PD-L1 expression was analyzed by IHC (Dako 22c3; PD-L1 positive: TPS >1%). TMB was measured by counting all somatic mutations found per tumor (TMB-high: >10 mutations/MB). Tumors with PAK4-high (H) and PAK4-low (L) expression were classified as those in top quartile and bottom 3 quartiles, respectively. Immune cell infiltrates were calculated by Quantiseq. Significance was determined using chi-square and Wilcoxon rank sum test and adjusted for multiple comparisons (q-value < 0.05). Survival was extracted from insurance claims data and calculated from the time of tissue collection to the last contact using Kaplan-Meier estimates. Results: There was no difference in median age, gender, smoking status, and histologic distribution between PAK-H and PAK-L tumors. Compared to PAK4-L, the PAK4-H was associated with higher frequency of co-mutations in TP53 (76.3% vs 63.9%, p < 0.0001) and RB1 (13.6% vs 8.1%, p < 0.0001). PAK4-H tumors were associated with greater genomic loss of heterozygosity (24.1% vs 16.4%, p < 0.0001), and expression of immune checkpoint genes ( LAG3, PDCD1, PDCD1LG2, CD274, IDO1, CTLA4, CD80, HAVCR2; p < 0.05). KRAS (including KRAS G12C) , BRAF, STK11, and EGFR mutations, and ALK and ROS1 fusions were less frequent in PAK4-H tumors (p < 0.001). A greater proportion of PAK4-H tumors were TMB-H (40.3% vs 33.3%, p < 0.0001) and PD-L1 negative (48.2% vs 40.8%, p < 0.001). PAK4-H tumors had lower infiltration of B cells, M1/M2 macrophages, CD8+ T-cells, and Tregs (p < 0.001). Overall survival (OS) was inferior among patients with PAK4-H tumors (median, 14.9 vs 21.5 months, HR = 1.28, 95% CI, 1.21-1.36, p < 0.001), which was driven by adenocarcinoma histology. Survival with immunotherapy was also inferior in patients with PAK4-H adenocarcinoma (median, 23.6 vs 30.3 months, HR = 1.23, 95% CI, 1.02-1.48, p = 0.03), but not squamous cell carcinoma. A similar trend in survival was noted in patients who received EGFR- and ALK inhibitors. Conclusions: PAK4-H expression in NSCLC is associated with a higher frequency of alterations predisposing to genomic instability, differentially modulated immune phenotype, and a lower frequency of actionable genomic alterations. Patients with PAK4-H adenocarcinoma have inferior OS, and survival with immunotherapy and targeted therapy. PAK4-H expression defines a subgroup of patients with un unmet need for novel treatment strategies.

Abstract PD6-06: Comprehensive characterization of neurotransmitters and neuronal signaling gene alterations in invasive breast cancers

BACKGROUNDIt has been reported that the sympathetic nervous system and associated neurotransmitters (NTs) play a pivotal role in driving breast cancer (BC) tumorigenesis and metastasis, however, comprehensive characterization of these pathways in BC is lacking. The purpose of this study was to retrospectively characterize NTs and neuronal signaling (NTNS) gene alterations in a large real-world BC cohort. METHODS A total of 6464 BC tumors were analyzed by next generation sequencing (NextSeq, 592 genes and WES, NovaSEQ, 720 genes) and whole transcriptome sequencing (WTS, NovaSeq) at Caris Life Sciences. Gene set variation analysis (GSVA) scores were calculated (positive: higher gene expressions in a selected gene set compared to genes outside that gene set in each tumor specimen, vice versa for negative) to assess expression of major NTNS genes, including GABA, nicotinic (NIC), muscarinic (MUS), dopamine (DA), reelin (RELN), and glial cell line-derived neurotrophic factor (GDNF). GSVA scores were compared by histologic subtype, primary or metastatic site, and hormone receptor (HR) and HER2 status with corrected Wilcoxon-Mann-Whitney testing. All significance levels were p&lt;0.01. RESULTS The 6464 BC specimens in this cohort included 2520 primary sites and 3944 metastasis (mets) (liver: 1012; lymph node: 714; bone: 575; lung: 420; brain: 196). Predictive biomarker status in this cohort was HR+/HER2-: 3705; HR+/HER2+: 238; HR-/HER2+: 189; TNBC: 1654. Invasive ductal carcinomas (IDC) were the most common histologic subtype and demonstrated significantly higher GSVA scores for RELN and NIC pathways with respect to invasive lobular carcinomas (ILC) (Table). TNBC tumors had significantly higher enrichment overall (GABA, -0.04 vs -0.14; RELN, -0.05 vs -0.31; DA, -0.03 vs -0.08; MUS, 0.13 vs -0.16; NIC, 0.01 vs -0.12; and GDNF, 0.04 vs -0.04). HR-/HER2+ had significantly higher scores in GABA, -0.04 vs -0.14; RELN, -0.03 vs -0.31; MUS, 0.12 vs -0.16; and NIC, -0.01 vs -0.12 genes. Brain mets had significantly enriched pathway scores for GABA, 0.30 vs -0.13; MUS, 0.15 vs -0.08; and NIC, 0.13 vs -0.09 compared to primary tumors. Similarly, GABA,0.09 vs -0.13; DA, 0.07 vs -0.05; MUS, 0.17 vs -0.08; and NIC, 0.02 vs -0.09 pathways were enriched in bone mets compared to those from primary tumors. CONCLUSION Our results demonstrate that NTNS pathways are significantly enriched in IDC, TNBC tumors, and particularly in brain and bone mets. Our data advance the current understanding of the role of NTNS pathways in BC tumorigenesis and metastasis. Further investigation on genetic. determinants and signaling alternations associated with the observed NTNS pathway deregulation is warranted and could inform the development of novel therapeutic strategies. Significant comparisons with Bonferroni corrected p values are shown with an asterisk.

Citation Format: Irene Kang, Krutika Deshpande, Sarah Persing, Jun Yin, Joanne Xiu, Wolfgang Michael Korn, Jia Zeng, Evanthia T Roussos-Torres, Janice Lu, Darcy Spicer, Stephen F Sener, Antoinette R Tan, Ashley Sumrall, David SB Hoon, Cynthia X Ma, Carey K Anders, Heather L McArthur, Reva Basho, Heinz-Josef Lenz, Josh Neman. Comprehensive characterization of neurotransmitters and neuronal signaling gene alterations in invasive breast cancers [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD6-06.

Comparative molecular profiling of pancreatic ductal adenocarcinoma (PDAC) of the head (H) versus body/tail (B/T) and the tumor immune microenvironment (TIME)

598

Background: PDAC of the H and B/T differ in embryonic origin, cell composition, blood supply, lymphatic and venous drainage, and innervation. H tumors tend cause symptoms earlier and to present at earlier stages compared to B/T cancers. The impact of PDAC tumor location on patient presentation and survival has been shown in large national data-based analyses, although with conflicting results. We aimed to compare the molecular and tumor immune microenvironment (TIME) profiles of PDAC of the H vs. B/T. Methods: A total of 3499 PDAC samples were analyzed via next-generation sequencing (NGS) of RNA (whole transcriptome, NovaSeq), DNA (NextSeq, 592 genes or NovaSeq, whole exome sequencing) and immunohistochemistry (IHC, Caris Life Sciences, Phoenix, AZ). RNA deconvolution was performed using QuantiSeq (Finotello 2019, Genome Medicine) to quantify the immune cell infiltration. Pathway gene enrichment analyses were done using Gene Set Enrichment Analysis (GSEA, Subramaniam 2015, PNAS). Significance was determined as p values adjusted for multiple correction (q) of < 0.05. Results: Anatomic subsites of PDAC tumors were grouped by primary tumor sites into H (N = 2058) or B/T (N = 1384). There were significantly more metastatic tumors profiled from H vs. B/T (57% vs. 44%, p < 0.001). KRAS mutations (93.8% vs. 90.2%), genomic loss of heterozygosity (12.7% vs. 9.1%), and several copy number alterations ( FGF3, FGF4, FGF19, CCND1, ZNF703, FLT4, MUTYH, TNFRS14) trended higher in B/T when compared to H (p < 0.05 but q > 0.05). GNAS mutations (2.2% vs. 0.7%) trended higher in H vs. B/T (p < 0.05). No significant difference in immuno-oncology (IO) markers (TMB, PD-L1, MSI-H) were observed, but expression analysis of IO-related genes showed significantly higher expression of CTLA4 and PDCD1 in H (q < 0.05, fold change 1.2 and 1.3) and IDO1 and PDCD1LG2 expression trended higher in B/T (p < 0.05, fold change 0.95). When comparing median cell abundance values as part of TIME analysis, H had increased immune infiltration of B cells (0.045 vs. 0.043), M2 macrophages (0.035 vs. 0.032), neutrophils (0.056 vs. 0.052), NK cells (0.027 vs. 0.026), CD8+ T cells (% > 0: 48.2% vs. 43.2%), while B/T had increased infiltration of M1 macrophages (0.035 vs. 0.032) (all q < 0.05). GSEA showed enrichment of CTLA4 (normalized enrichment score (NES) 1.6, false discovery rate (FDR) 0.19) and primary immunodeficiency pathway enrichment (NES 1.7, FDR 0.11) in H. Conclusions: To our knowledge, this is one of the largest cohort of PDAC tumors subjected to broad molecular profiling. Differences in IO-related gene expression and TIME cell distribution suggest that response to IO therapies may differ in PDAC arising from H vs B/T. Subtle differences in the genomic profliles of H vs. B/T tumors were also observed.

LRP1B and GRM3 expression in colorectal cancer

177

Background: LRP1B is a member of the low-density lipoprotein receptor family and a tumor suppressor found to be downregulated in colon cancer (CRC). GRM3 is a receptor of glutamate, an amino acid and neurotransmitter. Inhibition of GRM3 reduces CRC cell growth. Recent data from CALGB/SWOG 80405 suggests that mutations (MT) of either LRP1B or GRM3 are associated with better and worse overall survival (OS) in patients treated with bevacizumab (Bev), respectively. We investigate the association of LRP1B or GRM3 mRNA levels with outcomes. Methods: A total of 13,780 CRC tumors (male 7,497, female 6,283) underwent comprehensive molecular profiling (Caris Life Sciences). Analyses included next-generation sequencing of DNA (592 genes, NextSeq, WES, NovaSEQ) and RNA (NovaSeq). Significance with multiple correction was indicated with q, otherwise p value. Gene Set Enrichment Analyses (GSEA) were performed (significance p <.05). A Consensus Molecular Subtype (CMS) calling algorithm was developed using mRNA levels (transcripts per million; TPM). Time on treatment (TOT) with Bev was extracted from insurance claims. Results: Male patients had higher GRM3 expression (median TPM.55 vs..52, p <.001). GRM3 and LRP1B were both elevated in brain metastases (1.95 vs..40, q<.01;.53 vs..16, q<.01) and enriched in CMS4 subtype (both p <.001). Overexpression of GRM3 and LRP1B were significantly associated with MSS (.11 vs..07, p <.0001;.54 vs..39, p <.0001) and TMB low status (.11 vs..08, p <.0001;.54 vs..40, p <.0001). For MSS tumors, high LRP1B was associated with lower MT rates of APC (76% vs. 78%), KRAS (49% vs. 51%) and PIK3CA (15% vs. 17%). For MSI tumors, high LRP1B correlated with higher MT of MSH6 (41% vs. 32%), BRCA2 (28% vs. 20%) and PMS2 (12% vs. 6%). MSS tumors with high GRM3 had more APC (79% vs. 75%), less KRAS (47% vs. 52%) and SMAD4 (12% vs. 16%) MT and MSI with high GRM3 carried more APC (42% vs. 35%) and RAD50 (18% vs. 8%) MT. MSS tumors with low LRP1B showed upregulation of the EIF2 pathway while MTOR, RAB, and CDC42 pathways were enriched in MSI with low LRP1B. CDC42 and MTOR pathways were enriched in MSS tumors with low GRM3, and MSI with low GRM3 displayed enrichment of EIF2 and Notch pathways. In MSS tumors, both LRP1B and GRM3 were prognostic and associated with better survival (HR.66, 95% CI [.56-.78], p <.0001 for LRP1B; HR.79, 95% CI [.68-.92], p <.01 for GRM3) and high expression of either one was also associated with better prognosis for patients treated with Bev (HR.85, 95% CI [.70-.92], p <.01 for LRP1B; HR 0.88, 95% CI [.77-.99], p <.05 for GRM3). Conclusions: LRP1B and GRM3 appear to be important regulators in CRC because of their prognostic value and association with response to bevacizumab treatment. Both LRP1B and GRM3 are associated with pathways of cell cycle progression, cell migration, and DNA repair. A better understanding of their role in angiogenic signaling is critical to develop more effective strategies to improve response to bevacizumab or immunotherapy.

Identification of Novel CDH1-NRG2α and F11R-NRG2α Fusions in NSCLC Plus Additional Novel NRG2α Fusions in Other Solid Tumors by Whole Transcriptome Sequencing

Introduction

A novel CD74-NRG2α fusion has recently been identified in NSCLC. We surveyed a large tumor database comprehensively profiled by whole transcriptome sequencing to investigate the incidence and distribution of NRG2 fusions among various solid tumors.

Methods

Tumor samples submitted for clinical molecular profiling at Caris Life Sciences (Phoenix, AZ) that underwent whole transcriptome sequencing (NovaSeq [Illumina, San Diego, CA]) were retrospectively analyzed for NRG2 fusion events. All NRG2 fusions with sufficient reads (> three junctional reads spanning ≥ seven nucleotides) were identified for manual review, characterization of fusion class, intact functional domains, EGF-like domain isoforms, breakpoints, frame retention, and co-occurring alterations by next-generation sequencing (NextSeq [Illumina, San Diego, CA], 592 genes).

Results

Seven inframe functional (containing the intact EGF-like domain) NRG2α fusions were identified, namely, the following: (1) NSCLC (two of 9600, 0.02%: CDH1-NRG2α [C11, N2], F11R-NRG2α [F1, N4]); (2) endometrial (two of 3060, 0.065%: CPM-NRG2α [C2, N2], OPA3-NRG2α [O1, N2]); (3) ovarian (one of 5030, 0.02%: SPON1-NRG2α [S6, N2]); (4) prostate (one of 1600, 0.063%: PLPP1-NRG2α [P1, N2]); and (5) carcinoma of unknown origin (one of 1400, 0.07%: CYSTM1-NRG2α [C2, N2]). No NRG2β fusions were identified. Both NSCLC samples contained the reciprocal NRG2 fusions (NRG2-CDH1, NRG2-F11R). Almost all inframe NRG2α fusions have no (N = 6, 85.7%) or low (N = 1, 14.3%) programmed death-ligand 1 expression. No additional known driver mutations were identified in these seven NRG2α fusion-positive tumor samples.

Conclusions

Similar to NRG1 fusions, NRG2α fusions are recurrent and rare ligand-fusions in NSCLC and other multiple tumor types, especially gynecologic malignancies.

Assessment of immune biomarkers and establishing a triple negative phenotype in gynecologic cancers

OBJECTIVE

Immuno-oncology (IO) has rapidly evolved, with many IO therapies either approved or under investigation for multiple malignancies. Biomarkers exist that can predict response to IO therapies including PD-L1 expression, microsatellite instability (MSI), and total mutation burden (TMB). This paper serves to analyze the presence of these biomarkers across gynecologic cancers.

METHODS

A total of 16,300 gynecologic cancer specimens submitted for molecular profiling to Caris Life Sciences were reviewed. Immunohistochemistry was performed using the SP142 anti-PD-L1 clone and assessed for intensity. Next-generation sequencing, immunohistochemistry, and fragment analysis were used to determine MSI status. TMB was measured by counting all non-synonymous missense mutations found per tumor not previously described as germline alterations. Chi-Square, Fisher Exact, and the Kruskal-Wallis test were used to compare cohorts.

RESULTS

Of 16,300 specimens, 54.1% were ovarian, 37.2% uterine, 7.2% cervical, 0.3% vulvar, 1.2% vaginal, with 0.1% unspecified. MSI-H was most frequent in uterine cancer (17.7%) and only 1% of ovarian cancers. PD-L1 expression was present in 38.3% of cervical and 62.5% of vulvar cancers, but less than 8% of ovarian and uterine cancers. TMB-H was present in 21.1% cervical, 19.7% uterine, and 5% ovarian cancers. Few specimens exhibited a "triple positive" phenotype - 0.3% ovarian, 1.5% uterine, and 1.5% cervical. Associations were seen between MSI, TMB, and PD-L1 across all cancer types.

CONCLUSIONS

The frequency of individual biomarkers pertinent to IO therapy varies by cancer type. HPV-driven genital tract cancers have higher frequencies of PD-L1 expression, MSI-H, and TMBH. Endometrial cancers are characterized by MSI-H and TMB, whereas ovarian cancers have a low frequency of MSI-H and modest PD-L1 or TMBH. The incidence of 'triple positive" cases was less than 2%.

Abstract 2221: Whole transcriptome sequencing reveals oncogenic fusions in melanoma

Introduction: Oncogenic gene fusions are frequently identified in different cancers; however, the exact incidence of oncogenic fusions in melanoma is not well defined. Several targeted therapies are approved and considered the standard of care for patients whose solid or hematologic tumors harbor particular gene fusions, but their role as targets in melanoma also remains undelineated. We sought to determine the prevalence of oncogenic fusions in metastatic or locally advanced melanoma.

Methods: We retrospectively analyzed data from formalin-fixed paraffin-embedded (FFPE) tumor samples sent to a commercial CLIA-certified laboratory (Caris Life Sciences) from February 2019 to July 2020. Samples were profiled by next-generation sequencing of a 592-gene DNA panel, whole transcriptome sequencing, and immunohistochemistry (IHC).

Results: Melanoma specimens were analyzed from 1,255 subjects, of whom 478 (38.1%) were female and 777 (61.9%) were male, with a median age of 67 years. Of these specimens, 780 (63.1%) were from metastatic sites. We identified 33 (2.6%) cases with in-frame oncogenic fusions (14 novel) including 21 BRAF fusions and 4 RAF1 fusions, as well as fusions involving PRKCA (n=4), TERT (n=2), AXL (n=1), and FGFR3 (n=1). PD-L1 expression by IHC (SP142 or 28-8 antibody) was detected in 512 (42.5%) specimens, including 10 (32.3%) of the fusion-positive tumors, while 572 (47.4%) specimens were TMB-High (≥10 mutations/Mb), including 11 (33.3%) of the fusion-positive tumors, suggesting these patients may respond to immunotherapy. We identified 796 (63.4%) cases with RAS/RAF pathogenic or likely pathogenic mutations, including 373 (30.0%) BRAF p.V600X mutations. With the exception of a single BRAF p.S467L (Class 3, “kinase-dead”) mutation, the absence of BRAF mutations in BRAF fusion-positive tumors suggests these fusions are oncogenic drivers. However, tumors harboring PRKCA and TERT fusions were each detected with at least one MAPK pathway co-alteration (NRAS, NF1, or BRAF p.V600E mutation). Fusion transcripts with unknown pathogenicity were also detected in 668 (53.2%) cases. RNA expression analysis of key molecular pathways (including Wnt/β-catenin, PI3K/AKT/MTOR, DNA repair, INFG, and JAK/STAT) showed a high degree of variability among fusion-positive tumors and other mutational subgroups, which may reflect the heterogeneity common to melanoma, prior treatments or development of various resistance mechanisms.

Conclusions: Oncogenic gene fusions are rare in melanoma when compared with other genetic variants. We identified potentially actionable fusions as well as co-alterations in fusion-positive cases, with notably mutual exclusivity of BRAF fusions and p.V600X mutations. The data suggest that targetable variants, including oncogenic fusions, may be identified in melanoma with comprehensive tumor profiling and provide patients with personalized treatment or clinical trial options.

Citation Format: Sourat Darabi, Andrew Elliott, David R. Braxton, Jia Zeng, Kelsey Poorman, Jeffrey Swensen, Geoffrey T. Gibney, Justin C. Moser, Thuy Phung, Michael B. Atkins, Gino K. In, Wolfgang Michael Korn, Burton L. Eisenberg, Michael J. Demeure. Whole transcriptome sequencing reveals oncogenic fusions in melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2221.

Genomic and immunologic characterization of large-cell neuroendocrine carcinoma of the lung

8535

Background: Large-cell neuroendocrine carcinoma (LCNEC) is a rare type of lung cancer with a poor prognosis. Due to its rarity, molecular characterization of LCNEC is not well elucidated. We aim to understand the genomic and immunologic landscape of LCNEC to identify molecular alterations and relevant biological pathways with potential therapeutic value. Methods: Comprehensive profiling including whole exome sequencing (WES), next-generation sequencing (NGS), whole transcriptome sequencing (WTS), and immunohistochemistry (IHC) for PD-L1 was performed (Caris Life Sciences, Phoenix, AZ). Tumor mutational burden (TMB) was calculated based on somatic nonsynonymous mutations. LCNEC was categorized as small cell lung cancer (SCLC)-like LCNEC ( TP53/ RB1 co-mutated) and non-small-cell lung cancer (NSCLC)-like LCNEC (wild type for one or both of TP53/ RB1). Molecular features of LCNEC were compared among the subgroups and with those of SCLC using the χ2 test with Benjamini & Hochberg correction. Results: A total of 467 cases of LCNEC were included. Commonly altered genes (≥ 5%) included TP53 (79.1%), RB1 (36.8%), SMARCA4 (10.4%), ARID1A (10.3%), KRAS (9.7%), KEAP1 (9.2%), KMT2D (8.7%), STK11 (8.4%), NF1 (7.1%), PTEN (6.1%), and CDKN2A (5.9%) . The prevalence of potentially actionable mutations was as follows: EGFR exon 19 deletion (0.48%), EGFR L858R (0.48%), ALK fusion (1.7%), KRAS G12C (2.9%). EGFR exon 19 deletion, EGFR L858R, and ALK fusion were exclusive to NSCLC-like LCNEC tumors. RET fusion, NTRK fusion and BRAFV600E were not detected. Copy number alterations (CNAs) were found in MYC (8.8%), ZNF703 (4.1%), FOXA1 (4.0%), FGFR1 (4.0%), ATK2 (3.9%), CCNE1 (3.7%), FGF19 (3.4%), TNFRSF14 (3.4%), and CCND1 (2.7%). Over-expression of cMET was noted in 10% and PD-L1 expression (by 22C3 pharmDx) of > 1% was noted in 21.5% of samples. WTS detected cMET exon 14 skipping mutations in 2.4% of samples. High tumor mutation burden (TMB; ≥ 10 Mut/MB) was seen in 40.6%. Among the 467 cases of LCNEC, 112 (24%) were SCLC-like LCNEC and 335 (76%) NSCLC-like LCNEC. Mutations in KRAS (12%), STK11 (11%), CDKN2A (9%), and SMARCA4 (14%) were more common in NSCLC-like LCNEC, compared with SCLC-like LCNEC (p value < 0.05). 442 cases of SCLC were compared with LCNEC tumors. SLFN11:SLFN12 fusion events, detected by WTS, were exclusively seen in SCLC and were not seen in any of the LCNEC cases. Gene expression profiles revealed that 1) B cell infiltration was higher in SCLC-like LCNEC, compared with SCLC, and 2) NK and T cell infiltration was lower, but B-cell infiltration was higher in NSCLC-like LCNEC, compared with SCLC. Conclusions: LCNEC displays a broad pattern of genomic alterations that overlap in the SCLC-like subset with the classic alterations in SCLC. The distinct genomic alterations and transcriptomic profiles present opportunities for therapeutic targeting and inform a future framework for development of therapeutics for LCNEC.