High-confidence calling of normal epithelial cells allows identification of a novel stem-like cell state in the colorectal cancer microenvironment.

Single-cell analyses can be confounded by assigning unrelated groups of cells to common developmental trajectories. For instance, cancer cells and admixed normal epithelial cells could adopt similar cell states thus complicating analyses of their developmental potential. Here, we develop and benchmark CCISM (for Cancer Cell Identification using Somatic Mutations) to exploit genomic single nucleotide variants for the disambiguation of cancer cells from genomically normal non-cancer cells in single-cell data. We find that our method and others based on gene expression or allelic imbalances identify overlapping sets of colorectal cancer versus normal colon epithelial cells, depending on molecular characteristics of individual cancers. Further, we define consensus cell identities of normal and cancer epithelial cells with higher transcriptome cluster homogeneity than those derived using existing tools. Using the consensus identities, we identify significant shifts of cell state distributions in genomically normal epithelial cells developing in the cancer microenvironment, with immature states increased at the expense of terminal differentiation throughout the colon, and a novel stem-like cell state arising in the left colon. Trajectory analyses show that the new cell state extends the pseudo-time range of normal colon stem-like cells in a cancer context. We identify cancer-associated fibroblasts as sources of WNT and BMP ligands potentially contributing to increased plasticity of stem cells in the cancer microenvironment. Our analyses advocate careful interpretation of cell heterogeneity and plasticity in the cancer context and the consideration of genomic information in addition to gene expression data when possible.

Robust detection of clinically relevant features in single-cell RNA profiles of patient-matched fresh and formalin-fixed paraffin-embedded (FFPE) lung cancer tissue.

PURPOSE: Single-cell transcriptional profiling reveals cell heterogeneity and clinically relevant traits in intra-operatively collected patient-derived tissue. So far, single-cell studies have been constrained by the requirement for prospectively collected fresh or cryopreserved tissue. This limitation might be overcome by recent technical developments enabling single-cell analysis of FFPE tissue. METHODS: We benchmark single-cell profiles from patient-matched fresh, cryopreserved and archival FFPE cancer tissue. RESULTS: We find that fresh tissue and FFPE routine blocks can be employed for the robust detection of clinically relevant traits on the single-cell level. Specifically, single-cell maps of fresh patient tissues and corresponding FFPE tissue blocks could be integrated into common low-dimensional representations, and cell subtype clusters showed highly correlated transcriptional strengths of signaling pathway, hallmark, and clinically useful signatures, although expression of single genes varied due to technological differences. FFPE tissue blocks revealed higher cell diversity compared to fresh tissue. In contrast, single-cell profiling of cryopreserved tissue was prone to artifacts in the clinical setting. CONCLUSION: Our analysis highlights the potential of single-cell profiling in the analysis of retrospectively and prospectively collected archival pathology cohorts and increases the applicability in translational research.

AHRR and SFRP2 in primary versus recurrent high-grade serous ovarian carcinoma and their prognostic implication.

BACKGROUND: The aim of this study was to analyse transcriptomic differences between primary and recurrent high-grade serous ovarian carcinoma (HGSOC) to identify prognostic biomarkers. METHODS: We analysed 19 paired primary and recurrent HGSOC samples using targeted RNA sequencing. We selected the best candidates using in silico survival and pathway analysis and validated the biomarkers using immunohistochemistry on a cohort of 44 paired samples, an additional cohort of 504 primary HGSOCs and explored their function. RESULTS: We identified 233 differential expressed genes. Twenty-three showed a significant prognostic value for PFS and OS in silico. Seven markers (AHRR, COL5A2, FABP4, HMGCS2, ITGA5, SFRP2 and WNT9B) were chosen for validation at the protein level. AHRR expression was higher in primary tumours (p < 0.0001) and correlated with better patient survival (p < 0.05). Stromal SFRP2 expression was higher in recurrent samples (p = 0.009) and protein expression in primary tumours was associated with worse patient survival (p = 0.022). In multivariate analysis, tumour AHRR and SFRP2 remained independent prognostic markers. In vitro studies supported the anti-tumorigenic role of AHRR and the oncogenic function of SFRP2. CONCLUSIONS: Our results underline the relevance of AHRR and SFRP2 proteins in aryl-hydrocarbon receptor and Wnt-signalling, respectively, and might lead to establishing them as biomarkers in HGSOC.

Negative Hyperselection of Resistance Mutations for Panitumumab Maintenance in RAS Wild-Type Metastatic Colorectal Cancer (PanaMa Phase II Trial, AIO KRK 0212).

PURPOSE: We evaluated additional mutations in RAS wild-type (WT) metastatic colorectal cancer (mCRC) as prognostic and predictive biomarkers for the efficacy of added panitumumab to a 5-fluorouracil plus folinic acid (FU/FA) maintenance as pre-specified analysis of the randomized PanaMa trial. PATIENTS AND METHODS: Mutations (MUT) were identified using targeted next-generation sequencing (NGS; Illumina Cancer Hotspot Panel v2) and IHC. RAS/BRAF V600E/PIK3CA/AKT1/ALK1/ERBB2/PTEN MUT and HER2/neu overexpressions were negatively hyperselected and correlated with median progression-free survival (PFS) and overall survival (OS) since start of maintenance treatment, and objective response rates (ORR). Univariate/multivariate Cox regression estimated hazard ratios (HR) and 95% confidence intervals (CI). RESULTS: 202 of 248 patients (81.5%) of the full analysis set (FAS) had available NGS data: hyperselection WT, 162 (80.2%); MUT, 40 (19.8%). From start of maintenance therapy, hyperselection WT tumors were associated with longer median PFS as compared with hyperselection MUT mCRC (7.5 vs. 5.4 months; HR, 0.75; 95% CI, 0.52-1.07; P = 0.11), OS (28.7 vs. 22.2 months; HR, 0.53; 95% CI, 0.36-0.77; P = 0.001), and higher ORR (35.8% vs. 25.0%, P = 0.26). The addition of panitumumab to maintenance was associated with significant benefit in hyperselection WT tumors for PFS (9.2 vs. 6.0 months; HR, 0.66; 95% CI, 0.47-0.93; P = 0.02) and numerically also for OS (36.9 vs. 24.9 months; HR, 0.91; 95% CI, 0.61-1.36; P = 0.50), but not in hyperselection MUT tumors. Hyperselection status interacted with maintenance treatment arms in terms of PFS (P = 0.06) and OS (P = 0.009). CONCLUSIONS: Extended molecular profiling beyond RAS may have the potential to improve the patient selection for anti-EGFR containing maintenance regimens.

BannMI deciphers potential n-to-1 information transduction in signaling pathways to unravel message of intrinsic apoptosis.

Motivation: Cell fate decisions, such as apoptosis or proliferation, are communicated via signaling pathways. The pathways are heavily intertwined and often consist of sequential interaction of proteins (kinases). Information integration takes place on the protein level via n-to-1 interactions. A state-of-the-art procedure to quantify information flow (edges) between signaling proteins (nodes) is network inference. However, edge weight calculation typically refers to 1-to-1 interactions only and relies on mean protein phosphorylation levels instead of single cell distributions. Information theoretic measures such as the mutual information (MI) have the potential to overcome these shortcomings but are still rarely used. Results: This work proposes a Bayesian nearest neighbor-based MI estimator (BannMI) to quantify n-to-1 kinase dependency in signaling pathways. BannMI outperforms the state-of-the-art MI estimator on protein-like data in terms of mean squared error and Pearson correlation. Using BannMI, we analyze apoptotic signaling in phosphoproteomic cancerous and noncancerous breast cell line data. Our work provides evidence for cooperative signaling of several kinases in programmed cell death and identifies a potential key role of the mitogen-activated protein kinase p38. Availability and implementation: Source code and applications are available at: https://github.com/zuiop11/nn_info and can be downloaded via Pip as Python package: nn-info.