Subclass analysis of donor HLA-specific IgG in antibody-incompatible renal transplantation reveals a significant association of IgG4 with rejection and graft failure.

Donor HLA-specific antibodies (DSAs) can cause rejection and graft loss after renal transplantation, but their levels measured by the current assays are not fully predictive of outcomes. We investigated whether IgG subclasses of DSA were associated with early rejection and graft failure. DSA levels were determined pretreatment, at the day of peak pan-IgG level and at 30 days post-transplantation in eighty HLA antibody-incompatible kidney transplant recipients using a modified microbead assay. Pretreatment IgG4 levels were predictive of acute antibody-mediated rejection (P = 0.003) in the first 30 days post-transplant. Pre-treatment presence of IgG4 DSA (P = 0.008) and day 30 IgG3 DSA (P = 0.03) was associated with poor graft survival. Multivariate regression analysis showed that in addition to pan-IgG levels, total IgG4 levels were an independent risk factor for early rejection when measured pretreatment, and the presence of pretreatment IgG4 DSA was also an independent risk factor for graft failure. Pretreatment IgG4 DSA levels correlated independently with higher risk of early rejection episodes and medium-term death-censored graft survival. Thus, pretreatment IgG4 DSA may be used as a biomarker to predict and risk stratify cases with higher levels of pan-IgG DSA in HLA antibody-incompatible transplantation. Further investigations are needed to confirm our results.

Handling limited datasets with neural networks in medical applications: A small-data approach.

MOTIVATION: Single-centre studies in medical domain are often characterised by limited samples due to the complexity and high costs of patient data collection. Machine learning methods for regression modelling of small datasets (less than 10 observations per predictor variable) remain scarce. Our work bridges this gap by developing a novel framework for application of artificial neural networks (NNs) for regression tasks involving small medical datasets. METHODS: In order to address the sporadic fluctuations and validation issues that appear in regression NNs trained on small datasets, the method of multiple runs and surrogate data analysis were proposed in this work. The approach was compared to the state-of-the-art ensemble NNs; the effect of dataset size on NN performance was also investigated. RESULTS: The proposed framework was applied for the prediction of compressive strength (CS) of femoral trabecular bone in patients suffering from severe osteoarthritis. The NN model was able to estimate the CS of osteoarthritic trabecular bone from its structural and biological properties with a standard error of 0.85MPa. When evaluated on independent test samples, the NN achieved accuracy of 98.3%, outperforming an ensemble NN model by 11%. We reproduce this result on CS data of another porous solid (concrete) and demonstrate that the proposed framework allows for an NN modelled with as few as 56 samples to generalise on 300 independent test samples with 86.5% accuracy, which is comparable to the performance of an NN developed with 18 times larger dataset (1030 samples). CONCLUSION: The significance of this work is two-fold: the practical application allows for non-destructive prediction of bone fracture risk, while the novel methodology extends beyond the task considered in this study and provides a general framework for application of regression NNs to medical problems characterised by limited dataset sizes.