A new clinical prediction tool for 5-year kidney transplant outcome

Use of disease embedding technique to predict the risk of progression to end-stage renal disease

The accurate prediction of progression of Chronic Kidney Disease (CKD) to End Stage Renal Disease (ESRD) is of great importance to clinicians and a challenge to researchers as there are many causes and even more comorbidities that are ignored by the traditional prediction models. We examine whether utilizing a novel low-dimensional embedding model disease2disease (D2D) learned from a large-scale electronic health records (EHRs) could well clusters the causes of kidney diseases and comorbidities and further improve prediction of progression of CKD to ESRD compared to traditional risk factors. The study cohort consists of 2,507 hospitalized Stage 3 CKD patients of which 1,375 (54.8%) progressed to ESRD within 3 years. We evaluated the proposed unsupervised learning framework by applying a regularized logistic regression model and a cox proportional hazard model respectively, and compared the accuracies with the ones obtained by four alternative models. The results demonstrate that the learned low-dimensional disease representations from EHRs can capture the relationship between vast arrays of diseases, and can outperform traditional risk factors in a CKD progression prediction model. These results can be used both by clinicians in patient care and researchers to develop new prediction methods.

The combination of area under the curve and percentage change in estimated glomerular filtration rate predicts long-term outcome of kidney transplants

The development of surrogate markers for long-term outcomes of kidney transplant (KT) is a focus of attention. We examined the possibility of using a combination of the area under the curve of estimated glomerular filtration rate (eGFR) for 2 years (AUC_eGFR2yrs ) and percent change in eGFR between 1 and 2 years after KT (% change_eGFR1/2yr ) as a surrogate marker. We compared the predictive power of death-censored graft failure with various combinations. The combination of >2% vs ≤2% for % change_eGFR1/2yr and >1300 vs ≤1300 mL/min/month for AUC_eGFR2yr had the highest Harrell C-index (0.647; 95% confidence interval [95% CI], 0.604-0.690). The death-censored graft survival rate of the group with ≤2% change_eGFR1/2yr and ≤1300 mL/min/month AUC_eGFR2yr was significantly lower than those of other groups. The AUC/% change eGFR had comparable predictive power to the previously identified marker ≥30% decline in eGFR between years 1 and 3 after KT (≤-30% change_eGFR1/3yr ) (Harrell's C-index = 0.645 [95% CI 0.628-0.662] for ≤-30% change_eGFR1/3yr ). The proposed combination might be useful as a surrogate marker in KT trials because it requires a shorter surveillance period than the established marker while having comparable predictive power.

A dynamic model for predicting graft function in kidney recipients' upcoming follow up visits: A clinical application of artificial neural network

BACKGROUND

Predicting the function of transplanted kidneys would help clinicians in individualized medical interventions. We aimed to develop and validate a predictive tool for a future value of estimated glomerular filtration rate (eGFR) at upcoming visits.

METHODS

We used static and time-dependent covariates as inputs of artificial neural network based prediction models for predicting an eGFR value for an upcoming visit. We included 675 kidney recipients, who received transplant in the Urmia kidney transplant center in 2001-2013 and were longitudinally cared for in 2001-2017. The first 75% of records of longitudinal data of each patient were used to develop the prediction models and the remaining last 25% for evaluating its performance. Models' performances were evaluated by Mean Square Error (MSE) and Mean Absolute Error (MAE).

RESULTS

The development and validation datasets included 18,773 and 7038 records of historical data, respectively. The most accurate model included 3 static covariates of recipients' gender and donors' age and gender as well as 11 dynamic covariates of recipients including current age, time since transplant, serum creatinine, fasting blood sugar, weight and blood pressures available at each visit time. The performance of prediction models in the validation cohort was improved when history window of time dependent variables' recent values was increased from 1 to 10 (an MSE decline from 161 to 99).

CONCLUSIONS

Our best performed model is able to dynamically predict a future eGFR value for kidney recipients' upcoming visits. Integrating such a clinical tool into daily workflow of outpatient clinics can potentially support clinicians in optimal and individualized decision makings.

A Machine Learning Approach Using Survival Statistics to Predict Graft Survival in Kidney Transplant Recipients: A Multicenter Cohort Study

Accurate prediction of graft survival after kidney transplant is limited by the complexity and heterogeneity of risk factors influencing allograft survival. In this study, we applied machine learning methods, in combination with survival statistics, to build new prediction models of graft survival that included immunological factors, as well as known recipient and donor variables. Graft survival was estimated from a retrospective analysis of the data from a multicenter cohort of 3,117 kidney transplant recipients. We evaluated the predictive power of ensemble learning algorithms (survival decision tree, bagging, random forest, and ridge and lasso) and compared outcomes to those of conventional models (decision tree and Cox regression). Using a conventional decision tree model, the 3-month serum creatinine level post-transplant (cut-off, 1.65 mg/dl) predicted a graft failure rate of 77.8% (index of concordance, 0.71). Using a survival decision tree model increased the index of concordance to 0.80, with the episode of acute rejection during the first year post-transplant being associated with a 4.27-fold increase in the risk of graft failure. Our study revealed that early acute rejection in the first year is associated with a substantially increased risk of graft failure. Machine learning methods may provide versatile and feasible tools for forecasting graft survival.

Predicting Individual Renal Allograft Outcomes Using Risk Models with 1-Year Surveillance Biopsy and Alloantibody Data

The ability to predict outcomes for individual patients would be a significant advance for not only counseling, but also identifying those for whom interventions may be needed. The goals of this study were to validate an existing risk prediction score that incorporates easily obtainable clinical factors and determine if histologic findings at 1-year surveillance biopsy and/or serum donor-specific alloantibody status could improve predictability of graft loss by 5 years. We retrospectively studied 1465 adults who received a solitary kidney transplant between January of 1999 and December of 2008 and had sufficiently detailed 5-year follow-up data for modeling. In this cohort, the Birmingham risk model (incorporating recipient factors at 1 year, including age, sex, ethnicity, renal function, proteinuria, and prior acute rejection) predicted death-censored and overall graft survival (c statistics =0.84 and 0.78, respectively). The presence of glomerulitis or chronic interstitial fibrosis (g and ci scores by Banff, respectively) on 1-year biopsy specimens independently correlated with graft loss by 5 years. Adding these variables to the model for death-censored graft loss increased predictability (c statistic =0.90), improved calibration (ability to stratify risk from high to low), and reclassified risk of failure in 29% of patients. Adding the presence of donor-specific alloantibody at 1 year did not improve predictability or reclassification but did improve calibration marginally. We conclude that, at 1 year after kidney transplant, a risk model of graft survival that incorporates clinical factors and histologic findings at surveillance biopsy is highly predictive of individual risk and well calibrated.