Sliding Window INteraction Grammar (SWING): a generalized interaction language model for peptide and protein interactions.

The explosion of sequence data has allowed the rapid growth of protein language models (pLMs). pLMs have now been employed in many frameworks including variant-effect and peptide-specificity prediction. Traditionally, for protein-protein or peptide-protein interactions (PPIs), corresponding sequences are either co-embedded followed by post-hoc integration or the sequences are concatenated prior to embedding. Interestingly, no method utilizes a language representation of the interaction itself. We developed an interaction LM (iLM), which uses a novel language to represent interactions between protein/peptide sequences. Sliding Window Interaction Grammar (SWING) leverages differences in amino acid properties to generate an interaction vocabulary. This vocabulary is the input into a LM followed by a supervised prediction step where the LM's representations are used as features. SWING was first applied to predicting peptide:MHC (pMHC) interactions. SWING was not only successful at generating Class I and Class II models that have comparable prediction to state-of-the-art approaches, but the unique Mixed Class model was also successful at jointly predicting both classes. Further, the SWING model trained only on Class I alleles was predictive for Class II, a complex prediction task not attempted by any existing approach. For de novo data, using only Class I or Class II data, SWING also accurately predicted Class II pMHC interactions in murine models of SLE (MRL/lpr model) and T1D (NOD model), that were validated experimentally. To further evaluate SWING's generalizability, we tested its ability to predict the disruption of specific protein-protein interactions by missense mutations. Although modern methods like AlphaMissense and ESM1b can predict interfaces and variant effects/pathogenicity per mutation, they are unable to predict interaction-specific disruptions. SWING was successful at accurately predicting the impact of both Mendelian mutations and population variants on PPIs. This is the first generalizable approach that can accurately predict interaction-specific disruptions by missense mutations with only sequence information. Overall, SWING is a first-in-class generalizable zero-shot iLM that learns the language of PPIs.

De novo identification of CD4+ T cell epitopes.

CD4+ T cells recognize peptide antigens presented on class II major histocompatibility complex (MHC-II) molecules to carry out their function. The remarkable diversity of T cell receptor sequences and lack of antigen discovery approaches for MHC-II make profiling the specificities of CD4+ T cells challenging. We have expanded our platform of signaling and antigen-presenting bifunctional receptors to encode MHC-II molecules presenting covalently linked peptides (SABR-IIs) for CD4+ T cell antigen discovery. SABR-IIs can present epitopes to CD4+ T cells and induce signaling upon their recognition, allowing a readable output. Furthermore, the SABR-II design is modular in signaling and deployment to T cells and B cells. Here, we demonstrate that SABR-IIs libraries presenting endogenous and non-contiguous epitopes can be used for antigen discovery in the context of type 1 diabetes. SABR-II libraries provide a rapid, flexible, scalable and versatile approach for de novo identification of CD4+ T cell ligands from single-cell RNA sequencing data using experimental and computational approaches.

Defenestrated endothelium delays liver-directed gene transfer in hemophilia A mice.

Hemophilia A is an inherited bleeding disorder caused by defective or deficient coagulation factor VIII (FVIII) activity. Until recently, the only treatment for prevention of bleeding involved IV administration of FVIII. Gene therapy with adeno-associated vectors (AAVs) has shown some efficacy in patients with hemophilia A. However, limitations persist due to AAV-induced cellular stress, immunogenicity, and reduced durability of gene expression. Herein, we examined the efficacy of liver-directed gene transfer in FVIII knock-out mice by AAV8-GFP. Surprisingly, compared with control mice, FVIII knockout (F8TKO) mice showed significant delay in AAV8-GFP transfer in the liver. We found that the delay in liver-directed gene transfer in F8TKO mice was associated with absence of liver sinusoidal endothelial cell (LSEC) fenestration, which led to aberrant expression of several sinusoidal endothelial proteins, causing increased capillarization and decreased permeability of LSECs. This is the first study to link impaired liver-directed gene transfer to liver-endothelium maladaptive structural changes associated with FVIII deficiency in mice.

Revealing and IgG4 analysis to factor VIII in haemophilia-A patients with and without inhibitors.

INTRODUCTION: The acquisition of factor VIII inhibitors poses major management challenges for haemophilia A (HA) patients. Most (Factor VIII) inhibitors are immunoglobulin G4 (IgG4) and G1 (IgG1) subclasses, with IgG4 being the most prevalent. The Nijmegen Bethesda Assay (NBA) was used to quantify inhibitors. However, the requirement for a large sample volume, accompanying costs, and required technical expertise complicate NBA, particularly in developing countries. ELISA-based screening proved to be more viable in a resource-constrained scenario. AIM: This study aimed to standardise and evaluate an in-house IgG4 ELISA for the detection of haemophilia A inhibitors. METHODS: This study enrolled thirty HA patients with inhibitors, thirty three HA without inhibitors, and 33 healthy controls. Standardisation of in-house IgG4 ELISA was performed. The checkerboard method was employed to optimise plasma-derived Factor VIII concentrations (HEMOFIL M Baxalta US Inc.), sample dilutions, and anti-human IgG4-HRP conjugate (Southern Biotechnology, USA). The samples were evaluated three times, and the mean optical density (OD) was used to determine the cutoff. RESULTS: Using a cutoff OD (mean±2SD) of 0.502 in our in-house ELISA, we could differentiate healthy controls and HA without inhibitors from HA with inhibitors with 93.3 % sensitivity, 97.0 % specificity, 97 % NPV, and 93.3 % PPV, respectively. However, the accuracy was 95.83 %. The two-way mixed-effects model, interclass correlation (ICC) derived by Cronbach's Alpha was 0.912 (p = 0.001) and close to perfect agreement. CONCLUSIONS: IgG4 ELISA is an effective method for detecting neutralising or functionally significant FVIII inhibitors, particularly in resource-constrained settings, following which patients may be referred to referral laboratories for quantification of inhibitors.

Radiation therapy enhances immunotherapy response in microsatellite stable colorectal and pancreatic adenocarcinoma in a phase II trial.

Overcoming intrinsic resistance to immune checkpoint blockade for microsatellite stable (MSS) colorectal cancer (CRC) and pancreatic ductal adenocarcinoma (PDAC) remains challenging. We conducted a single-arm, non-randomized, phase II trial (NCT03104439) combining radiation, ipilimumab and nivolumab to treat patients with metastatic MSS CRC (n = 40) and PDAC (n = 25) with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. The primary endpoint was disease control rate (DCR) by intention to treat. DCRs were 25% for CRC (ten of 40; 95% confidence interval (CI), 13-41%) and 20% for PDAC (five of 25; 95% CI, 7-41%). In the per-protocol analysis, defined as receipt of radiation, DCR was 37% (ten of 27; 95% CI, 19-58%) in CRC and 29% (five of 17; 95% CI, 10-56%) in PDAC. Pretreatment biopsies revealed low tumor mutational burden for all samples but higher numbers of natural killer (NK) cells and expression of the HERVK repeat RNA in patients with disease control. This study provides proof of concept of combining radiation with immune checkpoint blockade in immunotherapy-resistant cancers.

Mixing-based inhibitor screening in haemophilia A: challenges in interpretation.

: Inhibitor development in haemophilia A patients is a dreaded complication of factor VIII (FVIII) replacement therapy. With increasing use of FVIII replacement therapy, there is an imperative need for cost-effective and standardized screening. To evaluate the efficacy of mixing-based inhibitor screening (MBIS) in the detection of FVIII inhibitors and to assess the best cut-off values for MBIS. Forty inhibitor positive and 40 inhibitor negative haemophilia A patients, diagnosed by standard criteria, with detailed clinical, haematological and on-demand treatment records were included. MBIS was evaluated in all 80 cases and a classical Bethesda assay and Nijmegen modification of Bethesda assay (NBA) were used as gold standards for inhibitor diagnosis. Classical Bethesda assay missed eight cases, most with low titres, which were confirmed by NBA. A systematic analysis of cut-offs for MBIS using a receiver operating characteristic curve fixed the cut-off at more than 5 s. MBIS detected 36 out of 40 inhibitor positive haemophilia A patients with a sensitivity, specificity, PPV and NPV of 90.0, 95, 94.7, 90.5%, respectively, whereas at the conventional cut-off of more than 10 s, MBIS detected only 25 of 40 cases with a low sensitivity of 62.5%. The likelihood ratio of a positive test was 11. The false-negative haemophilia A patients had low titres from 1.6 to 4.2 BU/ml. MBIS at a cut-off of 5 s can be considered as an effective screening test in low-resource situations. In clinical situations and in cases with clinical evidence of inhibitors we recommend that a direct NBA should be done.

Clinicopathological parameters influencing inhibitor development in patients with hemophilia A receiving on-demand therapy.

BACKGROUND: Development of inhibitors to transfused factor VIII in patients with hemophilia A continues to be a challenge for professionals involved in hemophilia care. The majority of patients in India receive 'on-demand' rather than prophylactic therapy. The present study was done to assess the prevalence of factor VIII inhibitors in patients with hemophilia A (PWHA) receiving 'on-demand' therapy in a North Indian population and to study the clinicopathological parameters influencing the development of inhibitors. METHODS: The study group comprised of 300 PWHA. Detailed clinical parameters, treatment history, bleeding profile including family history were recorded. Diagnosis of hemophilia A was confirmed by relevant coagulation tests. Inhibitors were screened using mixing based studies followed by quantification by Bethesda assay and Nijmegen modified Bethesda assay. Samples were collected from five cities in North India where a free supply of factor VIII was available and effectively used in three of these cities. RESULTS: In the 300 PWHA, disease phenotype was severe in 219 (73%), moderate in 62 (20.67%) and mild in 19 (6.34%), based on the factor VIII bioassay. Inhibitor prevalence was 9.6% (n = 29) and seen only in the severe phenotype. Inhibitor titers ranged from 0.8 to 108.8 BU/ml. A total of 12 PWHA had low and 17 had high titers. Correlation of various clinicopathological parameters in inhibitor-positive versus negative PWHA showed significant correlation with age at onset of disease, severity of disease, age at first exposure to treatment, annual factor intake (IU/kg/year), intense treatment episodes and bleeding manifestations like central nervous system bleed and hematuria. The total study sample had blood group B in 33.34% PWHA, followed by O (27.34%), A (24.34%) and AB (15%), however, in inhibitor-positive samples, significant inhibitor formation was associated with the ABO subtype A (19/29, 65.51%). CONCLUSIONS: Factor VIII inhibitor prevalence in PWHA receiving 'on-demand' therapy was 9.6%. Clinicopathological correlates of inhibitor development in such PWHA have been analyzed in this novel study.

Clinical relevance of PD-L1 expression in gallbladder cancer: a potential target for therapy.

AIMS: Programmed death-ligand 1 (PD-L1), a potential target for immune checkpoint inhibitors in various solid neoplasms, has been studied in very few cases of Gall Bladder Carcinoma (GBC). The current study aimed to evaluate PD-L1 expression at primary and metastatic sites of GBC, and its associations with standard prognostic clinicopathological parameters, as well as with overall survival. METHODS AND RESULTS: One hundred and seventy-four cases of GBC were evaluated for PD-L1 expression by the use of the SP263 clone in tissue microarrays. Clinicopathological characteristics and survival data were correlated with PD-L1 expression analysed at different cut-offs of ≥1%, ≥10% and ≥50% in tumour cells and tumour-infiltrating lymphocytes (TILs). The mean age of patients was 49.9 years, and the male/female ratio was 1:2.9. Of the cases, 73.6% presented with stage 3/4 disease. Tumour cells expressed PD-L1 in 23.0% of cases, and TILs expressed PD-L1 in 24.1% of cases. At a cut-off of 10%, 14.9% of cases expressed PD-L1, and at a cut-off of 50%, 7.5% of cases expressed PD-L1. Significant associations were seen between tumour proportion score and histological type (P = 0.004), histological grade (P = 0.004), nuclear grade (P = 0.008), nodal metastasis (P = 0.051), higher stage (P = 0.058), and TILs (P < 0.001). Tumour size, growth pattern, the presence of necrosis and lymphovascular emboli showed no significant associations with PD-L1 in tumour cells or TILs. In synchronous paired samples from primary and metastatic lymph nodes, discordantly higher PD-L1 expression was evident in lymph nodes. Overall survival was not associated with PD-L1 expression (P = 0.546). CONCLUSION: PD-L1 does not appear to be a prognostic marker or influence survival in GBC patients. However, PD-L1 expression occurs in one of four GBCs, supporting the future possibility of immune-modulation therapy to improve the dismal overall survival.