Roberta Buonincontri

PD-L1 marks a subset of melanomas with a shorter overall survival and distinct genetic and morphological characteristics

BACKGROUND Programmed cell death ligand 1 (PD-L1) is a cell surface molecule that plays a critical role in suppressing immune responses, mainly through binding of the PD-1 receptor on T lymphocytes. PD-L1 may be expressed by metastatic melanoma (MM). However, its clinical and biological significance remains unclear. Here, we investigated whether expression of PD-L1 in MM identifies a biologically more aggressive form of the disease, carrying prognostic relevance. PATIENTS AND METHODS PD-L1 expression was analyzed by immunohistochemistry using two different antibodies in primary tumors and paired metastases from 81 melanoma patients treated at a single institution. Protein expression levels were correlated with PD-L1 mRNA, BRAF mutational status and clinical outcome. PD-L1(+) and PD-L1(-) subsets of the A375 cell line were stabilized in vitro and compared using gene expression profiling and functional assays. Results were confirmed using xenograft models. RESULTS PD-L1 membrane positivity was detected in 30/81 (37%) of patients. By multivariate analysis, Breslow thickness and PD-L1 membrane positivity were independent risk factors for melanoma-specific death {PD-L1 5% cutoff [hazard ratio (HR) 3.92, confidence interval (CI) 95% 1.61-9.55 P < 0.003], PD-L1 as continuous variable (HR 1.03, 95% CI 1.02-1.04 P < 0.002)}. PD-L1 expression defined a subset of the BRAF-mutated A375 cell line characterized by a highly invasive phenotype and by enhanced ability to grow in xenograft models. CONCLUSIONS PD-L1 is an independent prognostic marker in melanoma. If confirmed, our clinical and experimental data suggest that PD-L1(+) melanomas should be considered a disease subset with distinct genetic and morpho-phenotypic features, leading to enhanced aggressiveness and invasiveness.

The Krüppel-like factor 2 transcription factor gene is recurrently mutated in splenic marginal zone lymphoma

The Kruppel-like factor 2 transcription factor gene is recurrently mutated in splenic marginal zone lymphoma

The enzymatic activities of CD38 enhance CLL growth and trafficking: implications for therapeutic targeting

The ecto-enzyme CD38 is gaining momentum as a novel therapeutic target for patients with hematological malignancies, with several anti-CD38 monoclonal antibodies in clinical trials with promising results. In chronic lymphocytic leukemia (CLL) CD38 is a marker of unfavorable prognosis and a central factor in the pathogenetic network underlying the disease: activation of CD38 regulates genetic pathways involved in proliferation and movement. Here we show that CD38 is enzymatically active in primary CLL cells and that its forced expression increases disease aggressiveness in a xenograft model. The effect is completely lost when using an enzyme-deficient version of CD38 with a single amino-acid mutation. Through the enzymatic conversion of NAD into ADPR (ADP-ribose) and cADPR (cyclic ADP-ribose), CD38 increases cytoplasmic Ca2+ concentrations, positively influencing proliferation and signaling mediated via chemokine receptors or integrins. Consistently, inhibition of the enzymatic activities of CD38 using the flavonoid kuromanin blocks CLL chemotaxis, adhesion and in vivo homing. In a short-term xenograft model using primary cells, kuromanin treatment traps CLL cells in the blood, thereby increasing responses to chemotherapy. These results suggest that monoclonal antibodies that block the enzymatic activities of CD38 or enzyme inhibitors may prove therapeutically useful.

Mutations in NOTCH1 PEST domain orchestrate CCL19-driven homing of chronic lymphocytic leukemia cells by modulating the tumor suppressor gene DUSP22

Even if NOTCH1 is commonly mutated in chronic lymphocytic leukemia (CLL), its functional impact in the disease remains unclear. Using CRISPR/Cas9-generated Mec-1 cell line models, we show that NOTCH1 regulates growth and homing of CLL cells by dictating expression levels of the tumor suppressor gene DUSP22. Specifically, NOTCH1 affects the methylation of DUSP22 promoter by modulating a nuclear complex, which tunes the activity of DNA methyltransferase 3A (DNMT3A). These effects are enhanced by PEST-domain mutations, which stabilize the molecule and prolong signaling. CLL patients with a NOTCH1-mutated clone showed low levels of DUSP22 and active chemotaxis to CCL19. Lastly, in xenograft models, NOTCH1-mutated cells displayed a unique homing behavior, localizing preferentially to the spleen and brain. These findings connect NOTCH1, DUSP22, and CCL19-driven chemotaxis within a single functional network, suggesting that modulation of the homing process may provide a relevant contribution to the unfavorable prognosis associated with NOTCH1 mutations in CLL.

SLAMF1 regulation of chemotaxis and autophagy determines CLL patient response.

Chronic lymphocytic leukemia (CLL) is a variable disease; therefore, markers to identify aggressive forms are essential for patient management. Here, we have shown that expression of the costimulatory molecule and microbial sensor SLAMF1 (also known as CD150) is lost in a subset of patients with an aggressive CLL that associates with a shorter time to first treatment and reduced overall survival. SLAMF1 silencing in CLL-like Mec-1 cells, which constitutively express SLAMF1, modulated pathways related to cell migration, cytoskeletal organization, and intracellular vesicle formation and recirculation. SLAMF1 deficiency associated with increased expression of CXCR4, CD38, and CD44, thereby positively affecting chemotactic responses to CXCL12. SLAMF1 ligation with an agonistic monoclonal antibody increased ROS accumulation and induced phosphorylation of p38, JNK1/2, and BCL2, thereby promoting the autophagic flux. Beclin1 dissociated from BCL2 in response to SLAMF1 ligation, resulting in formation of the autophagy macrocomplex, which contains SLAMF1, beclin1, and the enzyme VPS34. Accordingly, SLAMF1-silenced cells or SLAMF1(lo) primary CLL cells were resistant to autophagy-activating therapeutic agents, such as fludarabine and the BCL2 homology domain 3 mimetic ABT-737. Together, these results indicate that loss of SLAMF1 expression in CLL modulates genetic pathways that regulate chemotaxis and autophagy and that potentially affect drug responses, and suggest that these effects underlie unfavorable clinical outcome experienced by SLAMF1(lo) patients.

Adenosine signaling mediates hypoxic responses in the chronic lymphocytic leukemia microenvironment

The chronic lymphocytic leukemia (CLL) niche is a closed environment where leukemic cells derive growth and survival signals through their interaction with macrophages and T lymphocytes. Here, we show that the CLL lymph node niche is characterized by overexpression and activation of HIF-1α, which increases adenosine generation and signaling, affecting tumor and host cellular responses. Hypoxia in CLL lymphocytes modifies central metabolic pathways, protects against drug-driven apoptosis, and induces interleukin 10 (IL-10) production. In myeloid cells, it forces monocyte differentiation to macrophages expressing IRF4, IDO, CD163, and CD206, hallmarks of the M2 phenotype, which promotes tumor progression. It also induces IL-6 production and enhances nurturing properties. Low oxygen levels decrease T-cell proliferation, promote glycolysis, and cause the appearance of a population of PD-1+ and IL-10-secreting T cells. Blockade of the A2A adenosine receptor counteracts these effects on all cell populations, making leukemic cells more susceptible to pharmacological agents while restoring immune competence and T-cell proliferation. Together, these results indicate that adenosine signaling through the A2A receptor mediates part of the effects of hypoxia. They also suggest that therapeutic strategies to inhibit the adenosinergic axis may be useful adjuncts to chemotherapy or tyrosine kinase inhibitors in the treatment of CLL patients.

Abstract 241: Up-regulation of PD-L1 in melanoma determines resistance to BRAF and MEK inhibitors, induces a more aggressive phenotype and is partially mediated through post-transcriptional mechanisms involving miR-17-5p

Introduction: The therapy of metastatic melanoma (MM) was radically changed by the introduction of BRAF inhibitors (BRAFi). Even if highly effective in the short term, patients invariably develop resistance in the long term. For this reason other inhibitors as well as alternative or complementary therapeutic strategies are being tested in these patients. Among the immune checkpoint targets of clinical importance is PD-1, which is expressed by T cells and which binds to the PD-L1 ligand, which may be expressed by melanoma cells. We and others showed that PD-L1 is an independent negative prognostic marker for patients with MM. Methods: BRAFi-/MEKi-resistant melanoma cell lines were generated by treating cells with increasing concentrations of BRAFi or MEKi. Resistance, viability and aggressiveness were analyzed by MTT, migration and wound healing assays. Results were confirmed using xenograft models. Resistant cell lines were compared using RNAseq to identify enriched genetic pathways involved in the resistance. Luciferase reporter assay analysis was used to study the direct interactions between the PD-L1 and miR-17-5p. Results: By comparing responses to BRAFi in PD-L1+ and PD-L1- variants of the A375 cell line, we found that PD-L1 expression conferred resistance to BRAFi or MEKi. Conversely, silencing of the molecule restored sensitivity to these drugs. Resistant melanoma cell lines acquired PD-L1 expression and were characterized by a specific genetic profile, with the modulation of genes controlling cell movement and immune responses. Consistently, these cells showed a more aggressive behavior both in vitro and in xenograft models. PD-L1 silencing in resistant cells decreased invasive properties and restored expression of HLA-II molecules. PD-L1 up-regulation was only partly dependent on the paradoxical activation of the MAPK pathway, which characterized resistant cells. In addition, we found that resistance to BRAFi and MEKi down-modulated miR-17-5p, which showed an inverse correlation with PD-L1. Transfection of miR-17-5p into BRAFi-resistant cell lines induced the down-modulation of PD-L1, reduced the aggressive behavior of the cells and partially restored sensitivity to BRAFi. Finally, in the plasma of patients with MM, miR-17-5p was inversely correlated with expression of PD-L1 in the tumor tissue. Conclusions: These data demonstrate a direct link between expression of PD-L1 and resistance to BRAFi, as well as to a more aggressive behavior of melanoma cells. Furthermore, we define a novel post-transcriptional circuit responsible for PD-L1 up-regulation, based on a direct interaction between miR-17-5p and PD-L1 mRNA. Lastly, miR-17-5p plasmatic levels show an inverse correlation with PD-L1 expression by tumor cells, suggesting that they may be useful in monitoring disease outcome and drug sensitivity. Citation Format: Davide Brusa, Aureliano Stingi, Valentina Audrito, Francesca Orso, Sara Serra, Roberta Buonincontri, Francesco Neri, Gianna Baroni, Barbara Merelli, Romina Nassini, Daniela Massi, Salvatore Oliviero, Daniela Taverna, Mario Mandala, Silvia Deaglio. Up-regulation of PD-L1 in melanoma determines resistance to BRAF and MEK inhibitors, induces a more aggressive phenotype and is partially mediated through post-transcriptional mechanisms involving miR-17-5p. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 241.

Abstract 3192: Cooperation between adenosinergic and hypoxic signals in shaping chronic lymphocytic leukemia microenvironment

Extracellular adenosine generated from ATP/ADP through the concerted action of the ectoenzymes CD39 and CD73 elicits potent cytoprotective and immunosuppressive effects mediated by type-1 purinergic receptors. Chronic lymphocytic leukemia (CLL) patients expressing the ectoenzymes CD39 and CD73 can actively produce adenosine. This condition activates an autocrine adenosinergic axis that supports engraftment of leukemic cells in a growth-favorable environment. These effects are mediated by the A2A adenosine receptor, which inhibits chemotaxis and limits spontaneous and drug-induced apoptosis of CLL cells. Following the reported cross-talk between hypoxia and adenosine, we tested the hypothesis of a functional interplay between the adenosinergic axis and hypoxic signals in the CLL microenvironment. Results confirmed that CLL cells robustly increase HIF-1α expression when cultured under hypoxia. Under these conditions a significant increase in the mRNA and protein levels of CD73, CD26 and A2A was observed. An HPLC assay confirmed that hypoxic CLL cell cultures produce higher extracellular adenosine levels compared to normoxia. The increase is apparent upon inhibition of adenosine deaminase and nucleoside transporters, suggesting that under hypoxia scavenging mechanisms are activated. Attention was then focused on the stromal and T cell compartments, which are critical to the formation and maintenance of the leukemic niche. Hypoxia enhanced differentiation of circulating monocytes into nurse-like cells, macrophages of the M2 type playing an essential role in nurturing leukemic cells. During hypoxia, differentiated NLC overexpressed A2A, which was also functional. Indeed, an increased AKT and ERK1/2 phosphorylation was observed following pharmacological activation. The enhancement of NLC differentiation under hypoxic conditions relied, at least in part, on the activation of A2A: its engagement by the agonist favored NLC generation in normoxia, with overexpression of IDO, CD163 and CD206. Moreover, activation of A2A induced secretion of immunomodulatory cytokines, such as IL-6, IL-10 and CCL18, while pharmacological blockade of A2A under hypoxia prevented NLC growth. In the T cell compartment, hypoxic cultures were followed by the sharp up-regulation of the A2A, without significantly affecting the enzymes that generate adenosine. Co-cultures of T lymphocytes and CLL cells under hypoxia resulted in a dramatic decrease of T cell proliferation, partially rescued by antagonists of A2A. Furthermore, expression of the PD-1 immunoinhibitory receptor was enhanced in hypoxic T cells, suggesting that there are multiple inhibitory mechanisms. Together, these results indicate that the adenosinergic and hypoxic axes synergize in shaping the CLL niche, suggesting that the pharmacological inhibition of adenosinergic signals may counteract some of the effects mediated by an hypoxic environment. Citation Format: Sara Serra, Davide Brusa, Roberta Buonincontri, Valentina Audrito, Tiziana Vaisitti, Simon Robson, Silvia Deaglio. Cooperation between adenosinergic and hypoxic signals in shaping chronic lymphocytic leukemia microenvironment. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3192. doi:10.1158/1538-7445.AM2015-3192