Soluble C5b-9 As a Prognostic Biomarker for Thrombotic Microangiopathy at the Onset of Graft-Versus-Host Disease

Introduction: Thrombotic microangiopathy (TMA) is a known complication of allogeneic hematopoietic cell transplantation (HCT). Though the presence of graft-versus-host disease (GVHD) predicts the development of TMA, only a small subset of patients with GVHD will develop this condition. In the current study, we examined soluble C5b-9 (sC5b9), the terminal complement complex, as a potential biomarker for the development of TMA among patients with active GVHD.

Methods: We performed a nested case-control study using a prospective cohort of adult allogeneic HCT recipients transplanted during 2006-2013 at the Fred Hutchinson Cancer Research Center. Cases (TMA) with antecedent GVHD were ascertained and validated as previously described (BBMT 2019;25:570). Two controls (non-TMA) were randomly selected for each case at the time of TMA via the incidence density sampling method after matching on the onset timing and severity of prior GVHD. We tested plasma sC5b9 levels (exposure) in longitudinal samples at matched time points (pre-transplant, onset of GVHD, onset of TMA or matched time point) using a commercially available immunoassay kit. Furthermore, we reviewed patient records to determine the onset of systemic infection (including Gram negative bacteremia, invasive aspergillosis, BK viremia, HHV-6 invasive disease, CMV invasive disease, and EBV reactivation). Mean sC5b9 values and 95% confidence intervals were estimated at each time point. Conditional logistic regressions were used to estimate the association (odds ratio, OR) between the outcome (TMA vs. non-TMA) and the exposure (sC5b9 level) after accounting for matching. sC5b9 was modeled both as a continuous variable and a binary variable dichotomized at the median value.

Results: Among 208 adult allogeneic patients enrolled in the prospective cohort, we identified 38 patients (13 TMA cases and 25 non-TMA controls) with antecedent GVHD of similar time of onset and clinical grading. The median time to the onset of GVHD and TMA was 21 days (IQR 14-31) and 37 days (IQR 23-80), respectively. Six out of 13 cases developed TMA by 28 days. Mean sC5b9 levels were significantly higher in the TMA group than the non-TMA group at the onset of TMA vs. matched time point (377 vs. 248 ng/mL) and GVHD (369 vs. 240 ng/mL) but less so prior to transplant (243 vs. 200 ng/mL) (Figure 1). Furthermore, mean sC5b9 levels were elevated in patients experiencing active infection (364.10 ng/mL) and the values for each type of infection were shown in Table 2. When considered as a continuous variable, every 100 ng/mL increase in sC5b9 at the onset of GVHD was associated with an OR of 4.18 for TMA (p=0.02) (Table 1). As a binary variable, sC5b9 ≧ 250 ng/mL vs. < 250 ng/mL had an OR of 5.51 for TMA (p=0.04).

Conclusions: In the nested case-control study, we found that elevated sC5b9 level at the onset of acute GVHD was associated with the development of TMA after accounting for the timing and severity of GVHD. Previous studies have shown that testing of sC5b9 at day 28 was associated with TMA development; however, the prognostic value of day 28 testing was limited by the early onset of disease. Our study suggests that sC5b9 screening at the onset of GVHD may help predict which individuals would later develop TMA. It remains unclear whether complement activation is the driver or the effect of endothelial injury among GVHD patients. Finally, we caution that activation of the terminal complement pathway can also signify serious systemic infection. The interpretation of sC5b9 must be individualized according to the clinical scenario.

Acknowledgement: We would like to acknowledge and thank Kypha Inc for providing us with the immunoassay kits. We performed the study design and data analysis independently without industry funding.

View largeDownload slide

View largeDownload slide

Close modal