Bayesian optimization design for dose-finding based on toxicity and efficacy outcomes in phase I/II clinical trials.

In phase I trials, the main goal is to identify a maximum tolerated dose under an assumption of monotonicity in dose-response relationships. On the other hand, such monotonicity is no longer applied to biologic agents because a different mode of action from that of cytotoxic agents potentially draws unimodal or flat dose-efficacy curves. Therefore, biologic agents require an optimal dose that provides a sufficient efficacy rate under an acceptable toxicity rate instead of a maximum tolerated dose. Many trials incorporate both toxicity and efficacy data, and drugs with a variety of modes of actions are increasingly being developed; thus, optimal dose estimation designs have been receiving increased attention. Although numerous authors have introduced parametric model-based designs, it is not always appropriate to apply strong assumptions in dose-response relationships. We propose a new design based on a Bayesian optimization framework for identifying optimal doses for biologic agents in phase I/II trials. Our proposed design models dose-response relationships via nonparametric models utilizing a Gaussian process prior, and the uncertainty of estimates is considered in the dose selection process. We compared the operating characteristics of our proposed design against those of three other designs through simulation studies. These include an expansion of Bayesian optimal interval design, the parametric model-based EffTox design, and the isotonic design. In simulations, our proposed design performed well and provided results that were more stable than those from the other designs, in terms of the accuracy of optimal dose estimations and the percentage of correct recommendations.

Cutting Edge: Class II-like Structural Features and Strong Receptor Binding of the Nonclassical HLA-G2 Isoform Homodimer.

HLA-G is a natural tolerogenic molecule and has the following unique features: seven isoforms (HLA-G1 to HLA-G7), formation of disulfide-linked homodimers, and ß2-microglobulin (ß2m)-free forms. Interestingly, individuals null for the major isoform, HLA-G1, are healthy and expressed the α2 domain-deleted isoform, HLA-G2, which presumably compensates for HLA-G1 function. However, the molecular characteristics of HLA-G2 are largely unknown. In this study, we unexpectedly found that HLA-G2 naturally forms a ß2m-free and nondisulfide-linked homodimer, which is in contrast to the disulfide-bonded ß2m-associated HLA-G1 homodimer. Furthermore, single-particle analysis, using electron microscopy, revealed that the overall structure and domain organization of the HLA-G2 homodimer resemble those of the HLA class II heterodimer. The HLA-G2 homodimer binds to leukocyte Ig-like receptor B2 with slow dissociation and a significant avidity effect. These findings provide novel insights into leukocyte Ig-like receptor B2-mediated immune regulation by the HLA-G2 isoform, as well as the gene evolution of HLA classes.

Evaluation of the Reactivity and Receptor Competition of HLA-G Isoforms toward Available Antibodies: Implications of Structural Characteristics of HLA-G Isoforms.

The human leucocyte antigen (HLA)-G, which consists of seven splice variants, is a tolerogenic immune checkpoint molecule. It plays an important role in the protection of the fetus from the maternal immune response by binding to inhibitory receptors, including leukocyte Ig-like receptors (LILRs). Recent studies have also revealed that HLA-G is involved in the progression of cancer cells and the protection from autoimmune diseases. In contrast to its well characterized isoform, HLA-G1, the binding activities of other major HLA-G isoforms, such as HLA-G2, toward available anti-HLA-G antibodies are only partially understood. Here, we investigate the binding specificities of anti-HLA-G antibodies by using surface plasmon resonance. MEM-G9 and G233 showed strong affinities to HLA-G1, with a nM range for their dissociation constants, but did not show affinities to HLA-G2. The disulfide-linker HLA-G1 dimer further exhibited significant avidity effects. On the other hand, 4H84 and MEM-G1, which can be used for the Western blotting of HLA-G isoforms, can bind to native HLA-G2, while MEM-G9 and G233 cannot. These results reveal that HLA-G2 has a partially intrinsically disordered structure. Furthermore, MEM-G1, but not 4H84, competes with the LILRB2 binding of HLA-G2. These results provide novel insight into the functional characterization of HLA-G isoforms and their detection systems.

Impact of G29179T mutation on two commercial PCR assays for SARS-CoV-2 detection.

Nucleic acid amplification test (NAAT) is the gold standard for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection. However, genetic mutations in the virus can affect the result. Cycle threshold (Ct) values of N genes and their association with mutations using SARS-CoV-2 positive specimens diagnosed by the Xpert Xpress SARS-CoV-2 were examined in this study. In total, 196 nasopharyngeal swab specimens were tested for SARS-CoV-2 infection using the Xpert Xpress SARS-CoV-2, and 34 were positive. WGS was performed for four outlier samples with increased ΔCt identified by Scatterplot analysis as well as seven control samples without increased ΔCt in the Xpert Xpress SARS-CoV-2. The presence of the G29179T mutation was identified as a cause of increased ΔCt. PCR using the Allplex™ SARS-CoV-2 Assay did not show a similar increase in ΔCt. Previous reports focusing on N-gene mutations and their effects on SARS-CoV-2 testing including the Xpert Xpress SARS-CoV-2 were also summarized. While a single mutation that impacts one target of a multiplex NAAT is not a true detection failure, mutation compromising NAAT target region can cause confusion of the results and render the assay susceptible to diagnostic failure.

Bayesian optimization design for finding a maximum tolerated dose combination in phase I clinical trials.

The development of combination therapies has become commonplace because potential synergistic benefits are expected for resistant patients of single-agent treatment. In phase I clinical trials, the underlying premise is toxicity increases monotonically with increasing dose levels. This assumption cannot be applied in drug combination trials, however, as there are complex drug-drug interactions. Although many parametric model-based designs have been developed, strong assumptions may be inappropriate owing to little information available about dose-toxicity relationships. No standard solution for finding a maximum tolerated dose combination has been established. With these considerations, we propose a Bayesian optimization design for identifying a single maximum tolerated dose combination. Our proposed design utilizing Bayesian optimization guides the next dose by a balance of information between exploration and exploitation on the nonparametrically estimated dose-toxicity function, thereby allowing us to reach a global optimum with fewer evaluations. We evaluate the proposed design by comparing it with a Bayesian optimal interval design and with the partial-ordering continual reassessment method. The simulation results suggest that the proposed design works well in terms of correct selection probabilities and dose allocations. The proposed design has high potential as a powerful tool for use in finding a maximum tolerated dose combination.

Bayesian optimization for estimating the maximum tolerated dose in Phase I clinical trials.

We introduce a Bayesian optimization method for estimating the maximum tolerated dose in this article. A number of parametric model-based methods have been proposed to estimate the maximum tolerated dose; however, parametric model-based methods need an assumption that dose-toxicity relationships follow specific theoretical models. This assumption potentially leads to suboptimal dose selections if the dose-toxicity curve is misspecified. Our proposed method is based on a Bayesian optimization framework for finding a global optimizer of unknown functions that are expensive to evaluate while using very few function evaluations. It models dose-toxicity relationships with a nonparametric model; therefore, a more flexible estimation can be realized compared with existing parametric model-based methods. Also, most existing methods rely on point estimates of dose-toxicity curves in their dose selections. In contrast, our proposed method exploits a probabilistic model for an unknown function to determine the next dose candidate without ignoring the uncertainty of posterior while imposing some dose-escalation limitations. We investigate the operating characteristics of our proposed method by comparing them with those of the Bayesian-based continual reassessment method and two different nonparametric methods. Simulation results suggest that our proposed method works successfully in terms of selections of the maximum tolerated dose correctly and safe dose allocations.

Emergence of a multidrug-resistant plasmid encoding bla NDM-1, bla OXA-420 and armA in a clinical isolate of Acinetobacter variabilis in Japan.

Acinetobacter variabilis (formerly genospecies 15 sensu Tjernberg and Ursing) has been isolated from humans and animals and was proposed to be a novel species in 2015. A multidrug-resistant A. variabilis isolate, RYU24, was obtained in 2012 from an inpatient in Okinawa, Japan, with no record of overseas travel. The isolate was resistant to carbapenems, aminoglycosides and ciprofloxacin, with minimum inhibitory concentrations (MICs) of 32 µg ml-1 for imipenem and meropenem; > 1024 µg ml-1 for amikacin, arbekacin, gentamicin and tobramycin; and 8 µg ml-1 for ciprofloxacin. The isolate was found to harbour a 68-kbp plasmid carrying bla NDM-1, which encodes New Delhi metallo-ß-lactamase-1 (NDM-1); bla OXA-420, which encodes an OXA-58-like carbapenemase and; armA, which encodes ArmA 16S rRNA methylase conferring pan-aminoglycoside resistance. To our knowledge, this is the first report of a plasmid harbouring the three major drug-resistance genes, bla NDM-1, bla OXA-420 and armA.

Therapeutic application of human leukocyte antigen-G1 improves atopic dermatitis-like skin lesions in mice.

Human leukocyte antigen (HLA)-G is an immune checkpoint molecule that plays critical roles in immune response and in triggering inhibitory signaling to immune cells such as T cells, natural killer cells, and antigen-presenting cells. Thus, the application of HLA-G can be considered for treating immune response-related inflammatory disorders. We have previously reported that treatment with HLA-G1 and HLA-G2 ameliorates the joint swelling associated with collagen-induced arthritis of DBA/1 mice, an animal model for rheumatoid arthritis. In this study, we further investigated the effects of HLA-G1 on atopic dermatitis (AD), the most common inflammatory skin disorder. AD-like lesions were induced with the extract of the house dust mite Dermatophagoides farinae in NC/Nga mice. Continuous administration of HLA-G1 ameliorated the AD-like skin lesions in the mice. Furthermore, production of immunoglobulin E, interleukin (IL)-13, and IL-17A was significantly reduced in HLA-G1-treated mice, suggesting a Th2/Th17-mediated immune-inhibitory function of HLA-G1 in vivo. Our studies shed light on novel therapeutic strategies with recombinant HLA-G proteins for immune reaction-mediated chronic inflammatory disorders.

The immunosuppressive effect of domain-deleted dimer of HLA-G2 isoform in collagen-induced arthritis mice.

HLA-G is involved in maternal-fetal immune tolerance and is reported to be a natural tolerogenic molecule. Seven-spliced isoforms including dimeric and ß2m-free forms have been identified. The major isoform, HLA-G1 (and its soluble type HLA-G5), binds to the inhibitory immune receptors, leukocyte immunoglobulin (Ig)-like receptor (LILR) B1 and LILRB2. We previously reported that HLA-G1 also binds to paired Ig-like receptor (PIR)-B, a mouse homolog of LILRBs, and had a significant immunosuppressive effect in collagen-induced arthritis (CIA) mice. Although HLA-G2 and its soluble form HLA-G6 bind specifically to LILRB2, its functional characteristics are largely unknown. In this study, we report the significant immunosuppressive effect of HLA-G2 dimer in CIA mice. Surface plasmon resonance analysis revealed a specific interaction of HLA-G2 with PIR-B. CIA mice were administered HLA-G2 protein subcutaneously once in the left footpad and clinical severity was evaluated in a double-blind study. A single administration of HLA-G2 maintained a suppressive effect for over 1month. These results suggested that the HLA-G2 protein might be a useful biopharmaceutical for the treatment of rheumatoid arthritis by binding to inhibitory PIR-B.

TO901317, a potent LXR agonist, is an inverse agonist of CAR.

The basal transcriptional activity of unliganded human constitutive androstane receptor (hCAR) was shown to be repressed by the potent liver X receptor (LXR) agonist, TO901317, in a concentration-dependent manner using a reporter assay in cultured cells. TO901317 also repressed the basal transcriptional activity of both mouse and rat CAR. The certified hCAR agonist, CITCO, partially reversed this repressive effect of TO901317 on hCAR basal activity. Unlike hCAR, a three alanine insertion mutant and the splice variant 2 of hCAR require agonists, such as CITCO, to become transcriptionally active and the CITCO-induced reporter activity was repressed by TO901317. As has been previously shown for the typical hCAR inverse agonist, PK11195, TO901317 blocked the interaction of hCAR with steroid receptor co-activator 1 (SRC1). In contrast, the interaction between hCAR and nuclear receptor corepressor 1 (NCoR1) was promoted by PK11195 and TO901317. Furthermore, the hCAR-mediated basal induction of endogenous cytochrome P450 2B6 (CYP2B6) mRNA was adversely affected by co-treatment with TO901317.

The long-term immunosuppressive effects of disulfide-linked HLA-G dimer in mice with collagen-induced arthritis.

HLA-G, a natural immunosuppressant present in the human placenta during pregnancy, prevents fetal destruction by the maternal immune system. The immunosuppressive effect of HLA-G is mediated by the immune cell inhibitory receptors, LILRB1 and LILRB2. HLA-G forms disulfide-linked dimers by natural oxidation, and the dimer associates with LILRB1/B2 much more strongly than the monomer. Furthermore, the dimer formation remarkably enhanced the LILRB-mediated signaling. In this report, we studied the in vivo immunosuppressive effect of the HLA-G dimer, using the collagen-induced arthritis model mouse. Mice were treated with the HLA-G monomer or dimer intracutaneously at the left foot joint, once or for 5 days, and the clinical severity was evaluated daily in a double-blind study. The HLA-G monomer and dimer both produced excellent anti-inflammatory effects with a single, local administration. Notably, as compared to the monomer, the dimer exhibited significant immunosuppressive effects at lower concentrations, which persisted for about two months. In accordance with this result, a binding study revealed that the HLA-G dimer binds PIR-B, the mouse homolog of the LILRBs, with higher affinity and avidity than the monomer. The HLA-G dimer is expected to be quite useful as an anti-rheumatoid arthritis agent, in small amounts with minimal side effects.