Spatially-resolved transcriptomics reveal macrophage heterogeneity and prognostic significance in diffuse large B-cell lymphoma.

Macrophages are abundant immune cells in the microenvironment of diffuse large B-cell lymphoma (DLBCL). Macrophage estimation by immunohistochemistry shows varying prognostic significance across studies in DLBCL, and does not provide a comprehensive analysis of macrophage subtypes. Here, using digital spatial profiling with whole transcriptome analysis of CD68+ cells, we characterize macrophages in distinct spatial niches of reactive lymphoid tissues (RLTs) and DLBCL. We reveal transcriptomic differences between macrophages within RLTs (light zone /dark zone, germinal center/ interfollicular), and between disease states (RLTs/ DLBCL), which we then use to generate six spatially-derived macrophage signatures (MacroSigs). We proceed to interrogate these MacroSigs in macrophage and DLBCL single-cell RNA-sequencing datasets, and in gene-expression data from multiple DLBCL cohorts. We show that specific MacroSigs are associated with cell-of-origin subtypes and overall survival in DLBCL. This study provides a spatially-resolved whole-transcriptome atlas of macrophages in reactive and malignant lymphoid tissues, showing biological and clinical significance.

Patterns of Oncogene Coexpression at Single-Cell Resolution Influence Survival in Lymphoma.

Cancers often overexpress multiple clinically relevant oncogenes, but it is not known if combinations of oncogenes in cellular subpopulations within a cancer influence clinical outcomes. Using quantitative multispectral imaging of the prognostically relevant oncogenes MYC, BCL2, and BCL6 in diffuse large B-cell lymphoma (DLBCL), we show that the percentage of cells with a unique combination MYC+BCL2+BCL6- (M+2+6-) consistently predicts survival across four independent cohorts (n = 449), an effect not observed with other combinations including M+2+6+. We show that the M+2+6- percentage can be mathematically derived from quantitative measurements of the individual oncogenes and correlates with survival in IHC (n = 316) and gene expression (n = 2,521) datasets. Comparative bulk/single-cell transcriptomic analyses of DLBCL samples and MYC/BCL2/BCL6-transformed primary B cells identify molecular features, including cyclin D2 and PI3K/AKT as candidate regulators of M+2+6- unfavorable biology. Similar analyses evaluating oncogenic combinations at single-cell resolution in other cancers may facilitate an understanding of cancer evolution and therapy resistance. SIGNIFICANCE: Using single-cell-resolved multiplexed imaging, we show that selected subpopulations of cells expressing specific combinations of oncogenes influence clinical outcomes in lymphoma. We describe a probabilistic metric for the estimation of cellular oncogenic coexpression from IHC or bulk transcriptomes, with possible implications for prognostication and therapeutic target discovery in cancer. This article is highlighted in the In This Issue feature, p. 1027.

An ex vivo platform to guide drug combination treatment in relapsed/refractory lymphoma.

Although combination therapy is the standard of care for relapsed/refractory non-Hodgkin's lymphoma (RR-NHL), combination treatment chosen for an individual patient is empirical, and response rates remain poor in individuals with chemotherapy-resistant disease. Here, we evaluate an experimental-analytic method, quadratic phenotypic optimization platform (QPOP), for prediction of patient-specific drug combination efficacy from a limited quantity of biopsied tumor samples. In this prospective study, we enrolled 71 patients with RR-NHL (39 B cell NHL and 32 NK/T cell NHL) with a median of two prior lines of treatment, at two academic hospitals in Singapore from November 2017 to August 2021. Fresh biopsies underwent ex vivo testing using a panel of 12 drugs with known efficacy against NHL to identify effective single and combination treatments. Individualized QPOP reports were generated for 67 of 75 patient samples, with a median turnaround time of 6 days from sample collection to report generation. Doublet drug combinations containing copanlisib or romidepsin were most effective against B cell NHL and NK/T cell NHL samples, respectively. Off-label QPOP-guided therapy offered at physician discretion in the absence of standard options (n = 17) resulted in five complete responses. Among patients with more than two prior lines of therapy, the rates of progressive disease were lower with QPOP-guided treatments than with conventional chemotherapy. Overall, this study shows that the identification of patient-specific drug combinations through ex vivo analysis was achievable for RR-NHL in a clinically applicable time frame. These data provide the basis for a prospective clinical trial evaluating ex vivo-guided combination therapy in RR-NHL.

PLK1 inhibition selectively induces apoptosis in ARID1A deficient cells through uncoupling of oxygen consumption from ATP production.

Inhibitors of the mitotic kinase PLK1 yield objective responses in a subset of refractory cancers. However, PLK1 overexpression in cancer does not correlate with drug sensitivity, and the clinical development of PLK1 inhibitors has been hampered by the lack of patient selection marker. Using a high-throughput chemical screen, we discovered that cells deficient for the tumor suppressor ARID1A are highly sensitive to PLK1 inhibition. Interestingly this sensitivity was unrelated to canonical functions of PLK1 in mediating G2/M cell cycle transition. Instead, a whole-genome CRISPR screen revealed PLK1 inhibitor sensitivity in ARID1A deficient cells to be dependent on the mitochondrial translation machinery. We find that ARID1A knock-out (KO) cells have an unusual mitochondrial phenotype with aberrant biogenesis, increased oxygen consumption/expression of oxidative phosphorylation genes, but without increased ATP production. Using expansion microscopy and biochemical fractionation, we see that a subset of PLK1 localizes to the mitochondria in interphase cells. Inhibition of PLK1 in ARID1A KO cells further uncouples oxygen consumption from ATP production, with subsequent membrane depolarization and apoptosis. Knockdown of specific subunits of the mitochondrial ribosome reverses PLK1-inhibitor induced apoptosis in ARID1A deficient cells, confirming specificity of the phenotype. Together, these findings highlight a novel interphase role for PLK1 in maintaining mitochondrial fitness under metabolic stress, and a strategy for therapeutic use of PLK1 inhibitors. To translate these findings, we describe a quantitative microscopy assay for assessment of ARID1A protein loss, which could offer a novel patient selection strategy for the clinical development of PLK1 inhibitors in cancer.

p53-NEIL1 co-abnormalities induce genomic instability and promote synthetic lethality with Chk1 inhibition in multiple myeloma having concomitant 17p13(del) and 1q21(gain).

Recurrent cytogenetic abnormalities are the main hallmark of multiple myeloma (MM) and patients having 2 or more high-risk prognostic events are associated with extremely poor outcome. 17p13(del) and 1q21(gain) are critical and independent high-risk cytogenetic markers, however, the biological significance underlying the poor outcome in MM patients having co-occurrence of both these chromosomal aberrations has never been interrogated. Herein, we identified that patients harbouring concomitant 17p13(del) with 1q21(gain) demonstrated the worst prognosis as compared to patients with single- (either 17p13(del) or 1q21(gain)) and with no chromosomal events (WT for both chromosomal loci); and they are highly enriched for genomic instability (GI) signature. We discovered that the GI feature in the patients with concomitant 17p13(del)-1q21(gain) was recapitulating the biological properties of myeloma cells with co-existing p53-deficiency and NEIL1 mRNA-hyper-editing (associated with chromosome 17p and 1q, respectively) that have inherent DNA damage response (DDR) and persistent activation of Chk1 pathway. Importantly, this became a vulnerable point for therapeutic targeting whereby the cells with this co-abnormalities demonstrated hyper-sensitivity to siRNA- and pharmacological-mediated-Chk1 inhibition, as observed at both the in vitro and in vivo levels. Mechanistically, this was attributable to the synthetic lethal relationship between p53-NEIL1-Chk1 abnormalities. The Chk1 inhibitor (AZD7762) tested showed good synergism with standard-of-care myeloma drugs, velcade and melphalan, thus further reinforcing the translational potential of this therapeutic approach. In summary, combination of NEIL1-p53 abnormalities with an ensuing Chk1 activation could serve as an Achilles heel and predispose MM cells with co-existing 1q21(gain) and 17p13(del) to therapeutic vulnerability for Chk1 inhibition.

Machine-learning model derived gene signature predictive of paclitaxel survival benefit in gastric cancer: results from the randomised phase III SAMIT trial.

OBJECTIVE: To date, there are no predictive biomarkers to guide selection of patients with gastric cancer (GC) who benefit from paclitaxel. Stomach cancer Adjuvant Multi-Institutional group Trial (SAMIT) was a 2×2 factorial randomised phase III study in which patients with GC were randomised to Pac-S-1 (paclitaxel +S-1), Pac-UFT (paclitaxel +UFT), S-1 alone or UFT alone after curative surgery. DESIGN: The primary objective of this study was to identify a gene signature that predicts survival benefit from paclitaxel chemotherapy in GC patients. SAMIT GC samples were profiled using a customised 476 gene NanoString panel. A random forest machine-learning model was applied on the NanoString profiles to develop a gene signature. An independent cohort of metastatic patients with GC treated with paclitaxel and ramucirumab (Pac-Ram) served as an external validation cohort. RESULTS: From the SAMIT trial 499 samples were analysed in this study. From the Pac-S-1 training cohort, the random forest model generated a 19-gene signature assigning patients to two groups: Pac-Sensitive and Pac-Resistant. In the Pac-UFT validation cohort, Pac-Sensitive patients exhibited a significant improvement in disease free survival (DFS): 3-year DFS 66% vs 40% (HR 0.44, p=0.0029). There was no survival difference between Pac-Sensitive and Pac-Resistant in the UFT or S-1 alone arms, test of interaction p<0.001. In the external Pac-Ram validation cohort, the signature predicted benefit for Pac-Sensitive (median PFS 147 days vs 112 days, HR 0.48, p=0.022). CONCLUSION: Using machine-learning techniques on one of the largest GC trials (SAMIT), we identify a gene signature representing the first predictive biomarker for paclitaxel benefit. TRIAL REGISTRATION NUMBER: UMIN Clinical Trials Registry: C000000082 (SAMIT); ClinicalTrials.gov identifier, 02628951 (South Korean trial).

Quantitative imaging of RAD51 expression as a marker of platinum resistance in ovarian cancer.

Early relapse after platinum chemotherapy in epithelial ovarian cancer (EOC) portends poor survival. A-priori identification of platinum resistance is therefore crucial to improve on standard first-line carboplatin-paclitaxel treatment. The DNA repair pathway homologous recombination (HR) repairs platinum-induced damage, and the HR recombinase RAD51 is overexpressed in cancer. We therefore designed a REMARK-compliant study of pre-treatment RAD51 expression in EOC, using fluorescent quantitative immunohistochemistry (qIHC) to overcome challenges in quantitation of protein expression in situ. In a discovery cohort (n = 284), RAD51-High tumours had shorter progression-free and overall survival compared to RAD51-Low cases in univariate and multivariate analyses. The association of RAD51 with relapse/survival was validated in a carboplatin monotherapy SCOTROC4 clinical trial cohort (n = 264) and was predominantly noted in HR-proficient cancers (Myriad HRDscore < 42). Interestingly, overexpression of RAD51 modified expression of immune-regulatory pathways in vitro, while RAD51-High tumours showed exclusion of cytotoxic T cells in situ. Our findings highlight RAD51 expression as a determinant of platinum resistance and suggest possible roles for therapy to overcome immune exclusion in RAD51-High EOC. The qIHC approach is generalizable to other proteins with a continuum instead of discrete/bimodal expression.

Profiling of gastric cancer cell-surface markers to achieve tumour-normal discrimination.

BACKGROUND: Differentiating between malignant and normal cells within tissue samples is vital for molecular profiling of cancer using advances in genomics and transcriptomics. Cell-surface markers of tumour-normal discrimination have additional value in terms of translatability to diagnostic and therapeutic strategies. In gastric cancer (GC), previous studies have identified individual genes or proteins that are upregulated in cancer. However, a systematic analysis of cell-surface markers and development of a composite panel involving multiple candidates to differentiate tumour from normal has not been previously reported. METHODS: Whole transcriptome sequencing (WTS) of GC and matched normal samples from the Singapore Gastric Cancer Consortium (SGCC) was used as a discovery cohort to identify upregulated putative cell-surface proteins. Matched WTS data from the The Cancer Genome Atlas (TCGA) was used as a validation cohort. Promising candidates from this analysis were validated orthogonally using multispectral immunohistochemistry (mIHC) with automated quantitative analysis using the Vectra platform. mIHC was performed on a tissue microarray containing matched normal, marginal and tumour tissues. The receiver-operating characteristic (ROC) curves were analysed to identify markers with the highest diagnostic validity independently and in combination. RESULTS: Analysis of putative membrane protein transcripts from the SGCC discovery cohort WTS data (n=15 matched tumour and normal pairs) identified several differentially and highly expressed candidates in tumour compared with normal tissues. After validation with TCGA data (n=29 matched tumour and normal pairs), the following proteins were selected for mIHC analysis: CEACAM5, CEACAM6, CLDN4, CLDN7, and EpCAM. These were compared with established glycoprotein markers in GC, namely CA19-9 and CA72-4. Individual ROC curves yielded the best performance for CEACAM5 (area under the ROC curve (AUC)=0.80), CEACAM6 (AUC=0.82), EpCAM (AUC=0.83), and CA72-4 (AUC=0.76). Combined multiplexed imaging of these four markers revealed improved specificity and sensitivity for detection of tumour from normal tissue (AUC of 4-plex=0.91). CONCLUSION: CEAMCAM5, CEACAM6, EpCAM, and CA72-4 form a versatile set of markers for robust discrimination of GC from adjacent normal tissue. As cell-surface markers, they are compatible with both IHC and live imaging approaches. These candidates may be exploited to improve automated identification of tumour tissue in GC.

PRDM15 is a key regulator of metabolism critical to sustain B-cell lymphomagenesis.

PRDM (PRDI-BF1 and RIZ homology domain containing) family members are sequence-specific transcriptional regulators involved in cell identity and fate determination, often dysregulated in cancer. The PRDM15 gene is of particular interest, given its low expression in adult tissues and its overexpression in B-cell lymphomas. Despite its well characterized role in stem cell biology and during early development, the role of PRDM15 in cancer remains obscure. Herein, we demonstrate that while PRDM15 is largely dispensable for mouse adult somatic cell homeostasis in vivo, it plays a critical role in B-cell lymphomagenesis. Mechanistically, PRDM15 regulates a transcriptional program that sustains the activity of the PI3K/AKT/mTOR pathway and glycolysis in B-cell lymphomas. Abrogation of PRDM15 induces a metabolic crisis and selective death of lymphoma cells. Collectively, our data demonstrate that PRDM15 fuels the metabolic requirement of B-cell lymphomas and validate it as an attractive and previously unrecognized target in oncology.

The contribution of MYC and PLK1 expression to proliferative capacity in diffuse large B-cell lymphoma.

Polo-like kinase-1 (PLK1) regulates the MYC-dependent kinome in aggressive B-cell lymphoma. However, the role of PLK1 and MYC toward proliferation in diffuse large B-cell lymphoma (DLBCL) is unknown. We use multiplexed fluorescent immunohistochemistry (fIHC) to evaluate the co-localization of MYC, PLK1 and Ki67 to study their association with proliferation in DLBCL. The majority (98%, 95% CI 95-100%) of MYC/PLK1-double positive tumor cells expressed Ki67, underscoring the key role of the MYC/PLK1 circuit in proliferation. However, only 38% (95% CI 23-40%) and 51% (95% CI 46-51%) of Ki67-positive cells expressed MYC and PLK1, respectively. Notably, 40% (95% CI 26-43%) of Ki67-positive cells are MYC- and PLK-negative. A stronger correlation exists between PLK1 and Ki67 expression (R = 0.74, p < .001) than with MYC and Ki67 expression (R = 0.52, p < .001). Overall, the results indicate that PLK1 has a higher association than MYC in DLBCL proliferation and there are mechanisms besides MYC and PLK1 influencing DLBCL proliferation.

Multiplexed Fluorescent Immunohistochemical Staining, Imaging, and Analysis in Histological Samples of Lymphoma.

Immunohistochemical (IHC) methods for the in-situ analysis of protein expression by light microscopy are a powerful tool for both research and diagnostic purposes. However, the visualization and quantification of multiple antigens in a single tissue section using conventional chromogenic IHC is challenging. Multiplexed imaging is especially relevant in lymphoma research and diagnostics, where markers have to be interpreted in the context of a complex tumor microenvironment. Here we describe a protocol for multiplexed fluorescent IHC staining to enable the quantitative assessment of multiple targets in specific cell types of interest in lymphoma.The method covers aspects of antibody validation, antibody optimization, the multiplex optimization with markers of lymphoma subtypes, the staining of tissue microarray (TMA) slides, and the scanning of the slides, followed by data analysis, with specific reference to lymphoma. Using this method, scores for both the mean intensity of a marker of interest and the percentage positivity are generated to facilitate further quantitative analysis. Multiplexing minimizes sample utilization and provides spatial information for each marker of interest.

Biomarkers for Homologous Recombination Deficiency in Cancer.

Defective DNA repair is a common hallmark of cancer. Homologous recombination is a DNA repair pathway of clinical interest due to the sensitivity of homologous recombination-deficient cells to poly-ADP ribose polymerase (PARP) inhibitors. The measurement of homologous recombination deficiency (HRD) in cancer is therefore vital to the appropriate design of clinical trials incorporating PARP inhibitors. However, methods to identify HRD in tumors are varied and controversial. Understanding existing and new methods to measure HRD is important to their appropriate use in clinical trials and practice. The aim of this review is to summarize the biology and clinical validation of current methods to measure HRD, to aid decision-making for patient stratification and translational research in PARP inhibitor trials. We discuss the current clinical development of PARP inhibitors, along with established indicators for HRD such as germline BRCA1/2 mutation status and clinical response to platinum-based therapy. We then examine newer assays undergoing clinical validation, including 1) somatic mutations in homologous recombination genes, 2) "genomic scar" assays using array-based comparative genomic hybridization (aCGH), single nucleotide polymorphism (SNP) analysis or mutational signatures derived from next-generation sequencing, 3) transcriptional profiles of HRD, and 4) phenotypic or functional assays of protein expression and localization. We highlight the strengths and weaknesses of each of these assays, for consideration during the design of studies involving PARP inhibitors.

Epstein-Barr virus-associated primary nodal T/NK-cell lymphoma shows a distinct molecular signature and copy number changes.

The molecular biology of primary nodal T- and NK-cell lymphoma and its relationship with extranodal NK/T-cell lymphoma, nasal type is poorly understood. In this study, we assessed the relationship between nodal and extranodal Epstein-Barr virus-positive T/NK-cell lymphomas using gene expression profiling and copy number aberration analyses. We performed gene expression profiling and copy number aberration analysis on 66 cases of Epstein-Barr virus-associated T/NK-cell lymphoma from nodal and extranodal sites, and correlated the molecular signatures with clinicopathological features. Three distinct molecular clusters were identified with one enriched for nodal presentation and loss of 14q11.2 (TCRA loci). T/NK-cell lymphomas with a nodal presentation (nodal-group) were significantly associated with older age, lack of nasal involvement, and T-cell lineage compared to those with an extranodal presentation (extranodal-group). On multivariate analysis, nodal presentation was an independent factor associated with short survival. Comparing the molecular signatures of the nodal and extranodal groups it was seen that the former was characterized by upregulation of PD-L1 and T-cell-related genes, including CD2 and CD8, and downregulation of CD56, consistent with the CD8+/CD56-immunophenotype. PD-L1 and CD2 protein expression levels were validated using multiplexed immunofluorescence. Interestingly, nodal group lymphomas were associated with 14q11.2 loss which correlated with loss of TCR loci and T-cell origin. Overall, our results suggest that T/NK-cell lymphoma with nodal presentation is distinct and deserves to be classified separately from T/NK-cell lymphoma with extranodal presentation. Upregulation of PD-L1 indicates that it may be possible to use anti-PD1 immunotherapy in this distinctive entity. In addition, loss of 14q11.2 may be a potentially useful diagnostic marker of T-cell lineage.

Quantitative Analysis of a Multiplexed Immunofluorescence Panel in T-Cell Lymphoma.

Immunohistochemistry (IHC) provides clinically useful information on protein expression in cancer cells. However, quantification of colocalizing signals using conventional IHC and visual scores is challenging. Here we describe the application of quantitative immunofluorescence in angioimmunoblastic T-cell lymphoma (AITL), a peripheral T-cell lymphoma characterized by cellular heterogeneity that impedes IHC interpretation and quantification. A multiplexed immunofluorescence (IF) panel comprising T- and B-lymphocyte markers along with T-follicular helper (TFH) markers was validated for appropriate cellular localization in sections of benign tonsillar tissue and tested in two samples of AITL, using a Vectra microscope for spectral imaging and InForm software for analysis. We measured the percentage positivity of the TFH markers, BCL6 and PD1, in AITL CD4-positive cells to be approximately 26% and 45%, with 12% coexpressing both markers. The pattern is similar to CD4 cells within the germinal center of normal tonsils and clearly distinct from extragerminal CD4 cells. This study demonstrates the feasibility of automated and quantitative imaging of a multiplexed panel of cellular markers in formalin-fixed, paraffin-embedded tissue sections of a cellularly heterogenous lymphoma. Multiplexed IF allows the simultaneous scoring of markers in malignant and immune cell populations and could potentially increase accuracy for establishment of diagnostic thresholds.

Low Levels of NDRG1 in Nerve Tissue Are Predictive of Severe Paclitaxel-Induced Neuropathy.

INTRODUCTION: Sensory peripheral neuropathy caused by paclitaxel is a common and dose limiting toxicity, for which there are currently no validated predictive biomarkers. We investigated the relationship between the Charcot-Marie-Tooth protein NDRG1 and paclitaxel-induced neuropathy. METHODS/MATERIALS: Archived mammary tissue specimen blocks of breast cancer patients who received weekly paclitaxel in a single centre were retrieved and NDRG1 immunohistochemistry was performed on normal nerve tissue found within the sample. The mean nerve NDRG1 score was defined by an algorithm based on intensity of staining and percentage of stained nerve bundles. NDRG1 scores were correlated with paclitaxel induced neuropathy. RESULTS: 111 patients were studied. 17 of 111 (15%) developed severe paclitaxel-induced neuropathy. The mean nerve NDRG1 expression score was 5.4 in patients with severe neuropathy versus 7.7 in those without severe neuropathy (p = 0.0019). A Receiver operating characteristic (ROC) curve analysis of the mean nerve NDRG1 score revealed an area under the curve of 0.74 (p = 0.0013) for the identification of severe neuropathy, with a score of 7 being most discriminative. 13/54 (24%) subjects with an NDRG1 score < = 7 developed severe neuropathy, compared to only 4/57 (7%) in those with a score >7 (p = 0.017). CONCLUSION: Low NDRG1 expression in nerve tissue present within samples of surgical resection may identify subjects at risk for severe paclitaxel-induced neuropathy. Since nerve biopsies are not routinely feasible for patients undergoing chemotherapy for early breast cancer, this promising biomarker strategy is compatible with current clinical workflow.

CEACAM6 is upregulated by Helicobacter pylori CagA and is a biomarker for early gastric cancer.

Early detection of gastric cancers saves lives, but remains a diagnostic challenge. In this study, we aimed to identify cell-surface biomarkers of early gastric cancer. We hypothesized that a subset of plasma membrane proteins induced by the Helicobacter pylori oncoprotein CagA will be retained in early gastric cancers through non-oncogene addiction. An inducible system for expression of CagA was used to identify differentially upregulated membrane protein transcripts in vitro. The top hits were then analyzed in gene expression datasets comparing transcriptome of gastric cancer with normal tissue, to focus on markers retained in cancer. Among the transcripts enriched upon CagA induction in vitro, a significant elevation of CEACAM6 was noted in gene expression datasets of gastric cancer. We used quantitative digital immunohistochemistry to measure CEACAM6 protein levels in tissue microarrays of gastric cancer. We demonstrate an increase in CEACAM6 in early gastric cancers, when compared to matched normal tissue, with an AUC of 0.83 for diagnostic validity. Finally, we show that a fluorescently conjugated CEACAM6 antibody binds avidly to freshly resected gastric cancer xenograft samples and can be detected by endoscopy in real time. Together, these results suggest that CEACAM6 upregulation is a cell surface response to H. pylori CagA, and is retained in early gastric cancers. They highlight a novel link between CEACAM6 expression and CagA in gastric cancer, and suggest CEACAM6 to be a promising biomarker to aid with the fluorescent endoscopic diagnosis of early neoplastic lesions in the stomach.