Quantitative spatial evaluation of tumor-immune interactions in the immunotherapy setting of metastatic melanoma lymph nodes.

Introduction: Immune cell infiltration into the tumor microenvironment is generally associated with favorable clinical outcomes in solid tumors. However, the dynamic interplay among distinct immune cell subsets within the tumor-immune microenvironment as it relates to clinical responses to immunotherapy remains unresolved. In this study, we applied multiplex immunofluorescence (MxIF) to spatially characterize tumor-immune interactions within the metastatic melanoma lymph node. Methods: Pretreatment, whole lymph node biopsies were evaluated from 25 patients with regionally metastatic melanoma who underwent subsequent anti-PD1 therapy. Cyclic MxIF was applied to quantitatively and spatially assess expression of 45 pathologist-validated antibodies on a single tissue section. Pixel-based single cell segmentation and a supervised classifier approach resolved 10 distinct tumor, stromal and immune cell phenotypes and functional expression of PD1. Results: Single cell analysis across 416 pathologist-annotated tumor core regions of interest yielded 5.5 million cells for spatial evaluation. Cellular composition of tumor and immune cell subsets did not differ in the tumor core with regards to recurrence outcomes (p>0.05) however spatial patterns significantly differed in regional and paracrine neighborhood evaluations. Specifically, a regional community cluster comprised of primarily tumor and dendritic cells was enriched in patients that did not experience recurrence (p=0.009). By an independent spatial approach, cell-centric neighborhood analyses identified an enrichment for dendritic cells in cytotoxic T cell (CTL) and tumor cell-centric neighborhoods in the no recurrence patient response group (p<0.0001). Further evaluation of these neighborhoods identified an enrichment for CTL-dendritic cell interactions in patients that did not experience recurrence (p<0.0001) whereas CTL-macrophage interactions were more prevalent in CTL-centric neighborhoods of patients who experienced recurrence (p<0.0001). Discussion: Overall, this study offers a more comprehensive evaluation of immune infiltrates and spatial-immune signatures in the metastatic tumor-immune microenvironment as it informs recurrence risk following immunotherapy.

Resolving the heterogeneous tumor-centric cellular neighborhood through multiplexed, spatial paracrine interactions in the setting of immune checkpoint blockade.

Direct interactions between tumor and immune cells mediate the antitumor effect of all modern cancer immunotherapeutic agents. Simultaneously, tumor cells have evolved mechanisms of evasion including the downregulation of HLA-I potentially disrupting the mechanism of action employed by many immune checkpoint inhibitors. And yet the in situ interplay between these cells within the tumor immune microenvironment (TIME) remains elusive. Recent advances in histologic multiplex bioimaging platforms have enabled in-depth molecular characterization of single cells within spatially-preserved and clinically archived tumor tissues. Herein, we applied multiplex immunofluorescence (MxIF) to excisional lymph node biopsies from 14 patients with metastatic melanoma who experienced clear objective responses to immunotherapy (7 complete response; 7 progressive disease) to determine distinguishing features of the TIME in the pretreatment setting. Distinct regions of the TIME were evaluated using 35 proteins probing tumor, immune and vasculature components across 323 fields of view. Single cell compositional analysis confirmed established prognostic immune cell types including increased prevalence of cytotoxic T cells within the tumor core FOVs of responders. Integrating single cell quantification with the spatial arrangement of cellular neighborhoods surrounding tumor cells revealed novel, spatial immune signatures capable of stratifying TIME based on clinical response. Our analysis revealed dynamic cellular composition of the TCCN based on anatomical subregion, functional expression of HLA-I by the index tumor cell and ultimately clinical response to immunotherapy. Overall, this study provides an analytical framework to resolve the cellular complexity of the TIME, increasingly relevant to the outcomes of modern cancer immunotherapy.

Distinct Single Cell Gene Expression in Peripheral Blood Monocytes Correlates With Tumor Necrosis Factor Inhibitor Treatment Response Groups Defined by Type I Interferon in Rheumatoid Arthritis.

Previously, we demonstrated in test and validation cohorts that type I IFN (T1IFN) activity can predict non-response to tumor necrosis factor inhibitors (TNFi) in rheumatoid arthritis (RA). In this study, we examine the biology of non-classical and classical monocytes from RA patients defined by their pre-biologic treatment T1IFN activity. We compared single cell gene expression in purified classical (CL, n = 342) and non-classical (NC, n = 359) monocytes. In our previous work, RA patients who had either high IFNß/α activity (>1.3) or undetectable T1IFN were likely to have EULAR non-response to TNFi. In this study comparisons were made among patients grouped according to their pre-biologic treatment T1IFN activity as clinically relevant: "T1IFN undetectable (T1IFN ND) or IFNß/α >1.3" (n = 9) and "T1IFN detectable but IFNß/α ≤ 1.3" (n = 6). In addition, comparisons were made among patients grouped according to their T1IFN activity itself: "T1IFN ND," "T1IFN detected and IFNß/α ≤ 1.3," and "IFNß/α >1.3." Major differences in gene expression were apparent in principal component and unsupervised cluster analyses. CL monocytes from the T1IFN ND or IFNß/α >1.3 group were unlikely to express JAK1 and IFI27 (p < 0.0001 and p 0.0005, respectively). In NC monocytes from the same group, expression of IFNAR1, IRF1, TNFA, TLR4 (p ≤ 0.0001 for each) and others was enriched. Interestingly, JAK1 expression was absent in CL and NC monocytes from nine patients. This pattern most strongly associated with the IFNß/α>1.3 group. Differences in gene expression in monocytes among the groups suggest differential IFN pathway activation in RA patients who are either likely to respond or to have no response to TNFi. Additional transcripts enriched in NC cells of those in the T1IFN ND and IFNß/α >1.3 groups included MYD88, CD86, IRF1, and IL8. This work could suggest key pathways active in biologically defined groups of patients, and potential therapeutic strategies for those patients unlikely to respond to TNFi.

Plasma C5 glucose-to-2H2O ratio does not provide an accurate assessment of gluconeogenesis during hyperinsulinemic-euglycemic clamps in either nondiabetic or diabetic humans.

OBJECTIVE: Measurement of plasma C2 glucose enrichment is cumbersome. Therefore, the plasma C5 glucose-to-(2)H(2)O rather than the plasma C5-to-C2 glucose ratio commonly has been used to measure gluconeogenesis and glycogenolysis during hyperinsulinemic-euglycemic clamps. The validity of this approach is unknown. RESEARCH DESIGN AND METHODS: Ten nondiabetic and 10 diabetic subjects ingested (2)H(2)O the evening before study. The following morning, insulin was infused at a rate of 0.6 mU . kg(-1) . min(-1) and glucose was clamped at approximately 5.3 mmol/l for 5 h. Plasma C5 glucose, C2 glucose, and (2)H(2)O enrichments were measured hourly from 2 h onward. RESULTS: Plasma C2 glucose and plasma (2)H(2)O enrichment were equal in both groups before the clamp, resulting in equivalent estimates of gluconeogenesis and glycogenolysis. In contrast, plasma C2 glucose and plasma C5 glucose enrichments fell throughout the clamp, whereas plasma (2)H(2)O enrichment remained unchanged. Since the C5 glucose concentration and, hence, the C5 glucose-to-(2)H(2)O ratio is influenced by both gluconeogenesis and glucose clearance, whereas the C5-to-C2 glucose ratio is only influenced by gluconeogenesis, the C5 glucose-to-(2)H(2)O ratio overestimated (P < 0.01) gluconeogenesis during the clamp. This resulted in biologically implausible negative (i.e., calculated rates of gluconeogenesis exceeding total endogenous glucose production) rates of glycogenolysis in both the nondiabetic and diabetic subjects. CONCLUSIONS: Plasma C5 glucose-to-(2)H(2)O ratio does not provide an accurate assessment of gluconeogenesis in nondiabetic or diabetic subjects during a traditional (i.e., 2-3 h) hyperinsulinemic-euglycemic clamp. The conclusions of studies that have used this approach need to be reevaluated.

Comparison of the effects of pioglitazone and metformin on hepatic and extra-hepatic insulin action in people with type 2 diabetes.

OBJECTIVE: To determine mechanisms by which pioglitazone and metformin effect hepatic and extra-hepatic insulin action. RESEARCH DESIGN AND METHODS: Thirty-one subjects with type 2 diabetes were randomly assigned to pioglitazone (45 mg) or metformin (2,000 mg) for 4 months. RESULTS: Glucose was clamped before and after therapy at approximately 5 mmol/l, insulin raised to approximately 180 pmol/l, C-peptide suppressed with somatostatin, glucagon replaced at approximately 75 pg/ml, and glycerol maintained at approximately 200 mmol/l to ensure comparable and equal portal concentrations on all occasions. Insulin-induced stimulation of glucose disappearance did not differ before and after treatment with either pioglitazone (23 +/- 3 vs. 24 +/- 2 micromol x kg(-1) x min(-1)) or metformin (22 +/- 2 vs. 24 +/- 3 micromol x kg(-1) x min(-1)). In contrast, pioglitazone enhanced (P < 0.01) insulin-induced suppression of both glucose production (6.0 +/- 1.0 vs. 0.2 +/- 1.6 micromol x kg(-1) x min(-1)) and gluconeogenesis (n = 11; 4.5 +/- 0.9 vs. 0.8 +/- 1.2 micromol x kg(-1) x min(-1)). Metformin did not alter either suppression of glucose production (5.8 +/- 1.0 vs. 5.0 +/- 0.8 micromol x kg(-1) x min(-1)) or gluconeogenesis (n = 9; 3.7 +/- 0.8 vs. 2.6 +/- 0.7 micromol x kg(-1) x min(-1)). Insulin-induced suppression of free fatty acids was greater (P < 0.05) after treatment with pioglitazone (0.14 +/- 0.03 vs. 0.06 +/- 0.01 mmol/l) but unchanged with metformin (0.12 +/- 0.03 vs. 0.15 +/- 0.07 mmol/l). CONCLUSIONS: Thus, relative to metformin, pioglitazone improves hepatic insulin action in people with type 2 diabetes, partly by enhancing insulin-induced suppression of gluconeogenesis. On the other hand, both drugs have comparable effects on insulin-induced stimulation of glucose uptake.

Obesity and type 2 diabetes impair insulin-induced suppression of glycogenolysis as well as gluconeogenesis.

To determine whether the hepatic insulin resistance of obesity and type 2 diabetes is due to impaired insulin-induced suppression of glycogenolysis as well as gluconeogenesis, 10 lean nondiabetic, 10 obese nondiabetic, and 11 obese type 2 diabetic subjects were studied after an overnight fast and during a hyperinsulinemic-euglycemic clamp. Gluconeogenesis and glycogenolysis were measured using the deuterated water method. Before the clamp, when glucose and insulin concentrations differed among the three groups, gluconeogenesis was higher in the diabetic than in the obese nondiabetic subjects (P < 0.05) and glycogenolysis was higher in the diabetic than in the lean nondiabetic subjects (P < 0.05). During the clamp, when glucose and insulin concentrations were matched and glucagon concentrations were suppressed, both glycogenolysis and gluconeogenesis were higher (P < 0.01) in the diabetic versus the obese and lean nondiabetic subjects. Furthermore, glycogenolysis and gluconeogenesis were higher (P < 0.01) in the obese than in the lean nondiabetic subjects. Plasma free fatty acid concentrations correlated (P < 0.001) with glucose production and gluconeogenesis both before and during the clamp and with glycogenolysis during the clamp (P < 0.01). We concluded that defects in the regulation of glycogenolysis as well as gluconeogenesis cause hepatic insulin resistance in obese nondiabetic and type 2 diabetic humans.

Higher insulin concentrations are required to suppress gluconeogenesis than glycogenolysis in nondiabetic humans.

To determine the mechanism(s) by which insulin inhibits endogenous glucose production (EGP) in nondiabetic humans, insulin was infused at rates of 0.25, 0.375, or 0.5 mU. kg(-1). min(-1) and glucose was clamped at approximately 5.5 mmol/l. EGP, gluconeogenesis, and uridine-diphosphoglucose (UDP)-glucose flux were measured using [3-(3)H]glucose, deuterated water, and the acetaminophen glucuronide methods, respectively. An increase in insulin from approximately 75 to approximately 100 to approximately 150 pmol/l ( approximately 12.5 to approximately 17 to approximately 25 microU/ml) resulted in progressive (ANOVA; P < 0.02) suppression of EGP (13.1 +/- 1.3 vs. 11.7 +/- 1.03 vs. 6.4 +/- 2.15 micromol x kg(-1) x min(-1)) that was entirely due to a progressive decrease (ANOVA; P < 0.05) in the contribution of glycogenolysis to EGP (4.7 +/- 1.7 vs. 3.4 +/- 1.2 vs. -2.1 +/- 1.3 micro mol x kg(-1) x min(-1)). In contrast, both the contribution of gluconeogenesis to EGP (8.4 +/- 1.0 vs. 8.3 +/- 1.1 vs. 8.5 +/- 1.3 micro mol x kg(-1) x min(-1)) and UDP-glucose flux (5.0 +/- 0.4 vs. 5.0 +/- 0.3 vs. 4.0 +/- 0.5 micro mol x kg(-1) x min(-1)) remained unchanged. The contribution of the direct (extracellular) pathway to UDP-glucose flux was minimal and constant during all insulin infusions. We conclude that higher insulin concentrations are required to suppress the contribution of gluconeogenesis of EGP than are required to suppress the contribution of glycogenolysis to EGP in healthy nondiabetic humans. Since suppression of glycogenolysis occurred without a decrease in UDP-glucose flux, this implies that insulin inhibits EGP, at least in part, by directing glucose-6-phosphate into glycogen rather than through the glucose-6-phosphatase pathway.