Antibody-mediated rejection is a common cause of early kidney allograft loss but the specifics of antibody measurement, therapies and endpoints are undefined.
In this retrospective study, we assessed the performance of risk stratification using systematic donor-specific antibody (DSA) monitoring. Children who underwent kidney transplantation between 1/1/2010 and 3/1/2018 at Stanford with at least 12-month follow-up were included. Adverse graft outcome was defined by allograft failure, or decreased glomerular filtration rate (GFR), whichever occurred first. Allograft failure was defined as return to dialysis or preemptive re-transplantation. Decreased GFR was defined as estimated GFR <60 mL/min/ 1.73 m2 that persisted over at least 3 months. The creatinine-based “Bedside Schwartz” equation (2009) and/or Chronic Kidney Disease Epidemiology Collaboration equation were used to calculate GFR. 
A total of 233 patients were included with mean follow-up time of 45 (range, 12-108 months). Median age at transplant was 12.3 years, 46.8% were female, and 76% had a deceased donor transplant. Fifty-two (22%) formed C1q-binding de novo donor-specific antibodies (C1q-dnDSA). After a standardized augmented immunosuppressive protocol was implemented, C1q-dnDSA disappeared in 31 (58.5%).   All 233 patients were included in a random forest analysis. The most important characteristic for an adverse graft outcome (n=46) was the persistence of C1q binding.

The random forest model correctly classified only 17 of the 46 (37%) patients as having an adverse graft outcome while 184 of 187 (98.3%) were correctly classified as not having an event. While high immunodominant DSA (iDSA) mean fluorescence intensity (MFI) is a sensitive predictor of an adverse graft outcome, it did not usefully differentiate between patients who did well and those who did not. Looking at the iDSA MFI peak among the 118 patients who formed standard dnDSA, patients with adverse graft outcome had significantly higher iDSA MFI peak of 10861 (IQR 12394) compared to 2987 (IQR 6891) p=0.01. If we only analyze the C1q formers, there was no difference between MFI max in patients with or without adverse graft outcome, 13019 (IQR 13665) and 13569 (IQR 7421) respectively. Moreover, persistence of standard dnDSA in C1q-positive patients was not strongly correlated with graft failure, unlike C1q persistence. Seventeen out of 21 patients (80.9%) with C1q persistence had an adverse graft outcome. Of the 209 patients without C1q persistence, 29 (13.9%) had an adverse graft outcome.

Graft loss occurred in 16 patients at a median of 54 months (range 5-83), of whom 2 did not form dnDSA. The 14 patients who lost their graft due to rejection, all had persistent C1q-dnDSA. 
C1q binding status improves the individual risk assessment, with persistent C1q binding yielding the strongest independent association of graft loss (hazard ratio,45.46; 95% confidence interval, 11.7–177.4.) C1q-dnDSA is more useful than standard DSA as a noninvasive biomarker identifying patients at the highest risk of graft loss.