The data indicate that corneal epithelium contributes to elevated MMP9 and inflammatory cytokine expression in tears of KC patients. Cyclosporine A treatment reduced MMP9 and inflammatory cytokine levels in an in vitro inflammation model system. In KC patients, CyA treatment reduced MMP9 levels measured in tears with concomitant arrest of disease progression. Therefore, CyA might be a novel treatment strategy in KC patients but requires additional evaluation in larger cohorts. (ClinicalTrials.gov number, NCT01746823.).

GVHD (CGVHD) recently proposed criteria for diagnosis and assessment of overall CGVHD severity. We retrospectively reviewed 54 consecutive patients diagnosed with CGVHD between January 2002 and December 2005 after sibling donor transplant to assess the applicability of the new criteria in prognosticating survival and transplant-related mortality (TRM). A total of 8 patients (15%) were reclassified as late onset/ persistent or recurrent acute GVHD (late aGVHD), 15 (28%) had overlap syndrome and 31 (57%) had classic CGVHD. Three-year overall survival was worse in patients with late aGVHD (3-year probability 25% (95% CI 4-56%)) followed by overlap syndrome (3-year probability 87% (95% CI 56-96%)) and CGVHD (3-year probability 75% (95% CI 54-87%)); P ¼ 0.001. Among patients with overlap syndrome and CGVHD, a trend towards worse survival was seen in patients with severe disease (3-year probability 57.3% (95% CI 21-82%)) as compared to mild þ moderate disease (3-year probability 85.1% (95% CI 68-94)); P ¼ 0.1. This analysis, undertaken in a contemporary cohort of related donor recipients, indicates that the consensus guidelines are applicable to this population of CGVHD patients.

We present a comparative analysis of clinical presentation and response to treatment in 170 patients with chronic graft versus host disease (cGVHD) (123 following transplant from an unrelated donor [URD] and 47 from umbilical cord blood [UCB]). URD transplant recipients were significantly younger (median age 25 versus 39 years, P = .002; and the donor grafts were mostly HLA matched (67% versus 10%, P < .0001). UCB recipients had more frequent responses (complete remission [CR] + partial remission [PR]) to treatment (URD 48% versus UCB 74% at 2 months [P = .005]; 49% versus 78% at 6 months [P = .001] and 51% versus 72% at 1 year [P = .03] in the URD and UCB groups, respectively). Nonrelapse mortality (NRM) after diagnosis of cGVHD was worse after URD grafts. (1 year NRM 27% [19%-35%] URD versus 11% [2%-20%] UCB, P = .055). Separate multivariate analyses were performed in each cohort. In both, thrombocytopenia and no CR or PR at 2 months were independently associated with increased mortality. In addition, progressive onset of cGVHD was a significant predictor of increased mortality in URD cohort. These data suggest that cGVHD following UCB transplant may be more responsive to therapy and also lead to a lower NRM.

Citation: Russakoff DB, Mannil SS, Oakley JD, Ran AR, Cheung CY, Dasari S, Riyazzuddin M, Nagaraj S, Rao HL, Chang D, Chang RT. A 3D deep learning system for detecting referable glaucoma using full OCT macular cube scans. Trans Vis Sci Tech. 2020;9(2):12, https://doi.org/10. 1167/tvst.9.2.12 Purpose: The purpose of this study was to develop a 3D deep learning system from spectral domain optical coherence tomography (SD-OCT) macular cubes to differentiate between referable and nonreferable cases for glaucoma applied to real-world datasets to understand how this would affect the performance.Methods: There were 2805 Cirrus optical coherence tomography (OCT) macula volumes (Macula protocol 512 × 128) of 1095 eyes from 586 patients at a single site that were used to train a fully 3D convolutional neural network (CNN). Referable glaucoma included true glaucoma, pre-perimetric glaucoma, and high-risk suspects, based on qualitative fundus photographs, visual fields, OCT reports, and clinical examinations, including intraocular pressure (IOP) and treatment history as the binary (two class) ground truth. The curated real-world dataset did not include eyes with retinal disease or nonglaucomatous optic neuropathies. The cubes were first homogenized using layer segmentation with the Orion Software (Voxeleron) to achieve standardization. The algorithm was tested on two separate external validation sets from different glaucoma studies, comprised of Cirrus macular cube scans of 505 and 336 eyes, respectively. Results:The area under the receiver operating characteristic (AUROC) curve for the development dataset for distinguishing referable glaucoma was 0.88 for our CNN using homogenization, 0.82 without homogenization, and 0.81 for a CNN architecture from the existing literature. For the external validation datasets, which had different glaucoma definitions, the AUCs were 0.78 and 0.95, respectively. The performance of the model across myopia severity distribution has been assessed in the dataset from the United States and was found to have an AUC of 0.85, 0.92, and 0.95 in the severe, moderate, and mild myopia, respectively. Conclusions:A 3D deep learning algorithm trained on macular OCT volumes without retinal disease to detect referable glaucoma performs better with retinal segmentation preprocessing and performs reasonably well across all levels of myopia.Translational Relevance: Interpretation of OCT macula volumes based on normative data color distributions is highly influenced by population demographics and characteristics, such as refractive error, as well as the size of the normative database. Referable glaucoma, in this study, was chosen to include cases that should be seen by a specialist. This study is unique because it uses multimodal patient data for the glaucoma definition, and includes all severities of myopia as well as validates the algorithm with international data to understand generalizability potential.

The nature of chronic graft versus host disease (cGVHD) after umbilical cord blood (UCB) transplant has not been well characterized. We analyzed clinical presentation and response to treatment in 170 patients with cGVHD (123 following transplant from an unrelated donor (URD) and 47 following transplant from UCB). URD transplant recipients were significantly younger (median age 25 (0.5–57) versus 39 (0.2–61) years, p = 0.002; and were less likely to receive a transplant from an HLA mismatched donor (33% versus 90%, p&lt;0.0001). Seven% and 43% of patients underwent a transplant using a non myeloablative conditioning in the URD and UCB groups, respectively (p&lt; 0.001). Progressive onset of cGVHD was seen in 27% and 19% of patients in URD and UCB cohorts (p =0.15) and platelet count of &lt; 100,000/μl was seen in 36% and 15% of patients (p =0.014), respectively. The median follow up was 5 (0.14–10.8) and 1.5 (0.02–8.8) years in the 2 groups, respectively. A higher rate of response (CR+PR) to treatment was seen in UCB group across all time periods. Rates of response were 48% versus 74% at 2 months (p =0.005); 49% versus 78% at 6 months (p =0.001) and 51% versus 72% at 1 year (p =0.03) in the URD and UCB groups, respectively. Overall survival at 2 years was 59% (50–67%) versus 71% (53–83%) at 2 years and at 5 years was 54% (44–63%) versus 64% (42–79%) (p=0.17). In multivariate analysis, progressive onset of CGVHD (RR=2.3, 95% CI 1.4–3.8−2.3,p=0.002), platelet count of &lt;100,000/μl (RR=3.5, 95% CI 1.7–7.3,p=0.0009), liver involvement (RR=2.5, 95% CI 1.4–4.6,p=0.002), and no CR or PR at 2 months (RR=3.5, 95% CI 2.1–5.9,p&lt;0.0001) were associated with significantly increased mortality. These data suggests that among patients undergoing transplant from unrelated donor or cord blood, despite similar clinical presentation, higher rates of response to treatment were seen in UCB recipients. Early identification of high risk groups (progressive onset, less than PR at 2 months, platelet count &lt; 100,000/μl and liver involvement should facilitate timely assignment of intensified therapy. Response to Treatment Response 2 mo 2 mo 6 mo 6 mo 1 year 1year URD UCB URD UCB URD UCB CR+PR 49 (48) 29 (74) 54 (49) 32 (78) 53 (51) 23 (72) SD+Prog 53 (52) 10 (26) 56 (51) 9 (22) 52 (49) 9 (28)