Semaphorin-3A (sema3A) is a neuropilin-1 (np1) agonist. It inhibits the binding of the 165-amino acid form of VEGF (VEGF 165 ) to np1 and was reported to inhibit angiogenesis as a result. However, we find that sema3A concentrations that inhibit the mitogenic effects of VEGF 165 do not inhibit VEGF 165 -induced phosphorylation of VEGF receptor-2 (VEGFR-2). Furthermore, sema3A inhibits the biological effects of VEGF 121 , a VEGF form that does not bind to neuropilins and basic fibroblast growth factor, a growth factor whose activity, unlike that of VEGF, is not inhibited by small interfering RNA directed against np1. Therefore, the mechanism by which sema3A inhibits VEGF 165 activity does not depend on competition with VEGF 165 for binding to np1. Sema3A induced rapid disappearance of focal contacts followed by collapse of the actin cytoskeleton in human umbilical vein-derived endothelial cells. HEK293 cells expressing sema3A repel human endothelial cells and at high concentrations induce their death by apoptosis. Furthermore, sema3A inhibited the formation of tubes from endothelial cells in an in vitro angiogenesis assay. Similar effects are induced by the neuropilin-2 (np2) agonist sema3F. These inhibitory effects are abrogated by small interfering RNAs directed against np1 or np2, respectively. The anti-proliferative effects of sema3A and sema3F are additive when the semaphorins are added as pure proteins. However, when sema3A and sema3F were co-expressed in HEK293 cells their pro-apoptotic and cell repellant activities appeared to be synergistic. These observations suggest that combinations of sema3A and sema3F may be able to inhibit tumor angiogenesis more effectively than single semaphorins.

The class-3 semaphorins (sema3s) include seven family members. Six of them bind to neuropilin-1 (np1) or neuropilin-2 (np2) receptors or to both, while the seventh, sema3E, binds to the plexin-D1 receptor. Sema3B and sema3F were previously characterized as tumor suppressors and as inhibitors of tumor angiogenesis. To determine if additional class-3 semaphorins such as sema3A, sema3D, sema3E and sema3G possess anti-angiogenic and anti-tumorigenic properties, we expressed the recombinant full length semaphorins in four different tumorigenic cell lines expressing different combinations of class-3 semaphorin receptors. We show for the first time that sema3A, sema3D, sema3E and sema3G can function as potent anti-tumorigenic agents. All the semaphorins we examined were also able to reduce the concentration of tumor associated blood vessels although the potencies of the anti-angiogenic effects varied depending on the tumor cell type. Surprisingly, there was little correlation between the ability to inhibit tumor angiogenesis and their anti-tumorigenic activity. None of the semaphorins inhibited the adhesion of the tumor cells to plastic or fibronectin nor did they modulate the proliferation of tumor cells cultured in cell culture dishes. However, various semaphorins were able to inhibit the formation of soft agar colonies from tumor cells expressing appropriate semaphorin receptors, although in this case too the inhibitory effect was not always correlated with the anti-tumorigenic effect. In contrast, the anti-tumorigenic effect of each of the semaphorins correlated very well with tumor cell expression of specific signal transducing receptors for particular semaphorins. This correlation was not broken even in cases in which the tumor cells expressed significant concentrations of endogenous semaphorins. Our results suggest that combinations of different class-3 semaphorins may be more effective than single semaphorins in cases in which tumor cells express more than one type of semaphorin receptors.

Plexin-A4 is a receptor for sema6A and sema6B and associates with neuropilins to transduce signals of class-3 semaphorins. We observed that plexin-A1 and plexin-A4 are required simultaneously for transduction of inhibitory sema3A signals and that they form complexes. Unexpectedly, inhibition of plexin-A1 or plexin-A4 expression in endothelial cells using specific shRNAs resulted in prominent plexin type specific rearrangements of the actin cytoskeleton that were accompanied by inhibition of bFGF and VEGF-

Semaphorin-3B (sema3B) and semaphorin-3F (sema3F) are secreted tumor suppressors of lung cancer. Sema3F functions as an antiangiogenic factor that repels endothelial cells and compromises their proliferation/survival. However, tumor cells expressing either endogenous or recombinant sema3B fail to repel endothelial cells efficiently. Sema3B found in the conditioned medium of such cells is almost completely cleaved by furin-like pro-protein convertases, generating inactive 61-and 22-kDa fragments. We have generated a sema3B variant that was point mutated at the cleavage site (sema3B-m), thereby conferring partial resistance to cleavage. Conditioned medium from HEK293 cells expressing sema3b-m and conditioned medium of HEK293 cells expressing sema3B contained similar concentrations of semaphorin but sema3B-m was cleaved much less than sema3B. In contrast to HEK293 cells expressing native sema3B, cells expressing sema3b-m strongly repel endothelial cells. Conditioned medium from sema3B-mexpressing cells rapidly caused disassembly of focal adhesions and a collapse of the actin cytoskeleton of endothelial cells, inhibited vascular endothelial growth factor-induced phosphorylation of extracellular signal-regulated kinase 1/2, induced apoptosis of endothelial cells, and inhibited the formation of tubes from endothelial cells in an in vitro angiogenesis assay more potently than conditioned medium from cells expressing sema3B. Furthermore, HEK293 cells expressing sema3B-m inhibited basic fibroblast growth factorinduced angiogenesis in vivo much more potently than cells expressing sema3B. Repulsion of human umbilical vascular endothelial cells by sema3B-m was mediated primarily by the neuropilin-1 (np1) receptor but sema3B-m was also able to transduce signals via neuropilin-2 (np2). These results suggest that up-regulation of furin-like pro-protein convertases in malignant cells may enable tumors to evade the antiangiogenic effects of sema3B. [Cancer Res 2008;68(17):6922-31]

Immune checkpoint blockade (ICB) is an approved therapy for advanced metastatic mismatch repair (MMR)‐deficient cancer regardless of tissue of origin. Although therapy is effective initially, recurrence rates are significant, and long‐term outcomes remain poor for most patients. It is not currently recommended to give sequential ICB for advanced MMR‐deficient colorectal cancer (CRC) or for patients with metastatic cancer from Lynch syndrome. The need for subsequent therapy options in advanced MMR‐deficient cancer beyond the first ICB regimen arises in clinical practice, and there are often no effective standard chemotherapies or other targeted therapies. We report the case of a Lynch syndrome patient with metastatic CRC and urothelial cancer who was treated sequentially with pembrolizumab (targeting PD1), atezolizumab (targeting PD‐L1), brief rechallenge with pembrolizumab, and finally the combination of ipilimumab (targeting CTLA‐4) and nivolumab (targeting PD1). Over a 28‐month period the patient experienced prolonged disease control with each different regimen the first time it was given, including metabolic response by positron emission tomography and computed tomography scanning and tumor marker reductions. The case suggests that some patients with advanced MMR‐deficient CRC may experience meaningful clinical benefit from multiple sequential ICB regimens, a strategy that can be further tested in clinical trials. Key Points The case exemplifies clinical benefit from sequential immune checkpoint blockade in a patient with Lynch syndrome with advanced metastatic colorectal cancer and urothelial cancer. Metabolic response, with decreased fluorodeoxyglucose avidity on positron emission tomography and computed tomography, and reductions in tumor markers, such as carcinoembryonic antigen, were helpful in this case to monitor disease status over a 28‐month period of therapy. The concept of sequential immune checkpoint blockade in patients with advanced mismatch repair‐deficient cancer merits further study to determine which patients are most likely to benefit.

The neuropilins were originally characterized as cell membrane receptors that bind axon guidance factors belonging to the class-3 semaphorin subfamily. To transduce semaphorin signals, they form complexes with members of the plexin receptor family in which neuropilins serve as the ligand binding components and the plexins as the signal transducing components. The neuropilins were subsequently found to double as receptors for specific heparin binding splice forms of vascular endothelial growth factor (VEGF), and to be expressed on endothelial cells. This finding suggested that semaphorins may function as modulators of angiogenesis. It was recently found that several types of semaphorins such as semaphorin-3F function as inhibitors of angiogenesis while others, most notably semaphorin-4D, function as angiogenic factors. Furthermore, semaphorins such as semaphorin-3F and semaphorin-3B have been characterized as tumor suppressors and have been found to exert direct effects upon tumor cells. In this chapter we cover recent developments in this rapidly developing field of research.

Background: Liquid biopsy (LB) captures dynamic genomic alterations (alts) across metastatic colorectal cancer (mCRC) therapy and may complement tissue biopsy (TB). We sought to describe the utility of LB and better understand mCRC biology during therapy.Methods: Thirty-three patients (pts) with mCRC underwent LB. We used permutation-based t-tests to assess associations between alts, and clinical variables and used Kendall's tau to measure correlations.Results: Of 33 pts, 15 were women; 22 had colon, and the rest rectal cancer. Pts received a median of two lines of therapy before LB. Nineteen pts had limited testing on TB (RAS/RAF/TP53/APC), 11 extended NGS, and 3 no TB. Maxpct and alts correlated with CEA (p < 0.001, respectively). In 3/5 pts with serial LB, CEA correlated with maxpct trend, and CT tumor burden. In 6 pts, mutant RAS was seen in LB and not TB; 5/6 had received anti-EGFR therapy prior to LB, suggesting RAS alts developed post-therapy. In two pts RAS-mutated by TB, no RAS alts were detected on LB; these pts had low disease burden on CT at time of LB that also did not reveal APC or TP53 alts. In six patients who were KRAS wt based on TB, post anti-EGFR LB revealed subclonal KRAS mutations, likely a treatment effect. The median number of alts was higher post anti-EGFR LB (n = 12) vs. anti-EGFR naïve LB (n = 22) (9.5 vs. 5.5, p = 0.059) but not statistically significant. More alts were also noted in post anti-EGFR therapy LB vs. KRAS wt anti-EGFR-naïve LB (n = 6) (9.5 vs. 5) among patients with KRAS wild-type tumors, although the difference was not significant (p = 0.182).Conclusions: LB across mCRC therapy detects driver mutations, monitors disease burden, and identifies sub-clonal alts that reflect drug resistance, tumor evolution, and heterogeneity. Interpretation of LB results is impacted by clinical context.

Acute myeloid leukemia (AML) is an aggressive malignancy that requires rapid treatment with chemotherapies to reduce tumor burden. However, these chemotherapies can compromise lymphocyte function, thereby hindering normal anti-tumor immune responses and likely limiting the efficacy of subsequent immunotherapy. To better understand these negative impacts, we assessed the immunological effects of standard-of-care AML therapies on lymphocyte phenotype and function over time. When compared to healthy donors, untreated AML patients showed evidence of lymphocyte activation and exhaustion and had more prevalent CD57+NKG2C+ adaptive NK cells, which was independent of human cytomegalovirus (HCMV) status. HMA/venetoclax treatment resulted in a greater fraction of T cells with effector memory phenotype, inhibited IFN-γ secretion by CD8+ T cells, upregulated perforin expression in NK cells, downregulated PD-1 and 2B4 expression on CD4+ T cells, and stimulated Treg proliferation and CTLA-4 expression. Additionally, we showed increased expression of perforin and CD39 and enhanced IFN-γ production by T cells from pre-treatment blood samples of venetoclax-resistant AML patients. Our results provide insight into the lymphocyte status in previously untreated AML patients and the effects of standard-of-care treatments on their biology and functions. We also found novel pre-treatment characteristics of T cells that could potentially predict venetoclax resistance.