A Pilot Multi-Pronged Approach to Measuring Endothelial Dysfunction and Predict the Development of Endothelial Disorders and Death Post Hematopoietic Cell Transplantation (HCT)

Background: Endothelial dysfunction is a key driver of transplant associated thrombotic microangiopathy (TA-TMA) and sinusoidal obstructive syndrome (SOS)- early syndromes post hematopoietic cell transplantation, which are associated with non-relapse related mortality (NRM). Circulating endothelial cells (CEC), a marker of endothelial damage, are a diagnostic and prognostic biomarker in other TMAs, with scant data in TA-TMA and SOS. Non-invasive imaging modalities of endothelial health are also untested in both diseases. The primary objective of this pilot study was to determine the feasibility of routine hyperemia arterial tonometry (RH-PAT) and measuring circulating endothelial cells (CEC). Secondary objectives included determining the diagnostic, prognostic, and predictive performance of RH-PAT, CEC and endothelial markers and the development of TA-TMA, SOS and NRM.

Methods: In this prospective IRB approved study, patients were serially enrolled from April 2020- December 2022. Imaging and blood were obtained pre-HCT, day 0, day 14 and day 30. All ages were eligible for the blood sample component, but the RH-PAT arm was restricted to those >10 years. CEC were measured using flow cytometry (Farinacci et al. Res Pracr Thromb Haemost. 2019). ST2 and VCAM-1 were measured by ELISAs per manufacturing recommendations. RH-PAT is an FDA approved, non-invasive technique to measure endothelial health and calculates a reactive hyperemic index (lnRHI), lower lnRHI values are indicative or poor endothelial function. Descriptive statistics were used to compare groups.

Results: Fifty-one patients enrolled in the study, 21 participated in the RH-PAT arm, 43 in the blood sample arm, and 17 in both. Attrition rates were high in the RH-PAT arm, 4 people tolerated 1 study then declined additional tests, 5 underwent 2, and 12 tolerated 3 or more measurements. The median age at HCT was 10.1 years, 24 (49%) were female and 47 (96%) underwent allogeneic HCT. Ten (25%) were Black and 13 (27%) Hispanic.

Nineteen (37.2%) developed TA-TMA a median of 35 days post HCT (range 24.5 to 59). There were no significant differences in CEC at any time points in patients with TA-TMA compared to those without. LnRHI values were significantly lower pre-HCT in those who later developed TA-TMA compared to those who did not develop TA-TMA (mean 0.32, 95% CI 0.25-0.39 vs 0.61, 95% CI 0.44-0.78 respectively), though there were no differences on day 0, 14 or at the time of TA-TMA diagnosis.

Eleven patients developed SOS a median of 13 days post HCT (range 9.5 to 23). Seven patients experienced NRM a median of 134 days post HCT (range 69 to 353.5). CEC were not different among patients with SOS or those who died of NRM at any time point. Throughout the HCT process, CEC were present in very small amounts and peaked on day 14 (median of 0.04). LnRHI was significantly lower on day 30 in those who ultimately died of NRM versus those who did not (mean 0.25, 95% CI 0.13 to 0.36 vs 0.50, 95% CI 0.41 to 0.58 respectively). Day 30 lnRHI predicted NRM with an AUC of 0.81 (95% CI 0.59 to 1). However, there were no differences in lnRHI at other time points, nor were there any differences in lnRHI in those with SOS compared to no SOS at any time points. ST2 and sVCAM-1 were significantly elevated on day 14 and 30 in patients with TA-TMA, SOS and NRM. While endothelial markers ST2 and VCAM-1 had a significant positive correlation with each other, there were no significant correlations of these markers with CEC nor lnRHI, nor were CEC or lnRHI associated with each other.

Discussion: In this pilot study, we tested new imaging and blood biomarkers in children post HCT. CEC were low at all time points, difficult to isolate, and not different in those with endothelial disorders, SOS and TA-TMA. While small numbers, RH-PAT lnRHI was significantly lower pre-HCT preparative regimen in those who developed TA-TMA, which could suggest these patients had pre-existing endothelial injury, a potential risk factor for subsequent disease. At day 30, the RH-PAT lnRHI was associated with NRM,, perhaps suggesting that endothelial dysfunction of any etiology is a risk factor for death. While noninvasive, RH-PAT was poorly tolerated, limiting feasibility for future studies. While ST2 and sVCAM-1 were elevated on day 14 and day 30 in TA-TMA, SOS and NRM, they were non-specific. Additional studies are needed to identify alternative approaches of measuring endothelial dysfunction to predict and prognose TA-TMA and SOS.