Prospective external validation of biomarkers to predict acute graft-versus-host disease severity.

REGULAR ARTICLE Prospective external validation of biomarkers to predict acute graft-versus-host disease severity Marie Robin,1 Raphael Porcher,2,3 David Michonneau,1,4 Laetitia Taurines,1 Flore Sicre de Fontbrune,1 Ali
enor Xhaard,1 5 1 1 1,4,
lien Sutra Del Galy, Re
gis Peffault de Latour, Ge
rard Socie
, * and Marie-He
le ne Schlageter6,* Bastien Oriano, Aure 1 Service d’H
ematologie-Greffe, H^ opital Saint-Louis, APHP, Universit
e de Paris-Cit
e, Paris, France; 2CRESS, INSERM, INRAE, Paris, France; and 3Centre d’Epid
emiologie opital Saint-Louis, APHP, and 6Biologie Cellulaire, H^ opital Saint-Louis, U1131 INSERM, IRS, Universit
e Clinique, H^ opital H^ otel-Dieu, APHP, 4INSERM U976, 5Dermatologie, H^ de Paris-Cit
e, Paris, France Key Points   Biomarker panels at GVHD onset, independently from clinical parameters, were associated with survival and nonrelapse mortality. Using 3 different biostatistical tools, biomarkers only slightly improved prediction over clinical parameters. Acute graft-versus-host disease (GVHD) is still the major contributor to comorbidities and mortality after allogeneic hematopoietic stem cell transplantation. The use of plasmatic biomarkers to predict early outcomes has been advocated in the past decade. The purpose of this prospective noninterventional study was to test the ability of panels including 7 biomarkers (Elafin, HGF, IL2RA, IL8, REG3, ST2, and TNFRI), to predict day 28 (D28) complete response to steroid, D180 overall survival, and D180 nonrelapse mortality (NRM). Using previous algorithms developed by the Ann Arbor/MAGIC consortium, 204 patients with acute GVHD were prospectively included and biomarkers were measured at GVHD onset for all of them. Initial GVHD grade and bilirubin level were significantly associated with all those outcomes. After adjustment on clinical variables, biomarkers were associated with survival and NRM. In addition to clinical variables, biomarkers slightly improved the prediction of overall survival and NRM (concordance and net reclassification indexes). The potential benefit of adding biomarkers panel to clinical parameters was also investigated by decision curve analyses. The benefit of adding biomarkers to clinical parameters was however marginal for the D28 nonresponse and mortality endpoints. Introduction Hematopoietic stem cell transplantation (HSCT) remains the only curative treatment for many hematological diseases. HSCT’s efficacy relies on its immune effect against the hematological malignancy referred as the graft-versus-leukemia effect. Unfortunately, this graft-versus-leukemia is also closely related to acute graft-versus-host disease (GVHD), which induces tissue damage, immune defect, and eventually mortality. Acute GVHD occurs in approximately 50% of patients. Age, type of donor, sex, and HLA matching between donor and recipient, type of regimen, and GVHD prophylaxis influence the risk of acute GVHD. Clinical and histological grades of acute GVHD are predictive for outcome.1-5 Acute GVHD first-line therapy usually consists of high-dose corticosteroids. Unfortunately, up to 50% of patients will not achieve a sustainable complete remission (CR) with steroids and will be at higher risk of mortality. The main risk factor of GVHD-related mortality is the nonresponsiveness to steroids,6 but response is obviously unknown at GVHD onset. Finding biomarkers able to better predict acute GVHD severity and response to corticosteroids would thus be highly useful to personalize patient management. Submitted 4 March 2022; accepted 26 May 2022; prepublished online on Blood Advances First Edition 6 June 2022; final version published online 17 August 2022. DOI 10.1182/ bloodadvances.2022007477. *G.S. and M.-H.S. are joint last authors. Contact the corresponding author for data sharing: . 23 AUGUST 2022 • VOLUME 6, NUMBER 16 The full-text version of this article contains a data supplement. © 2022 by The American Society of Hematology. Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NCND 4.0), permitting only noncommercial, nonderivative use with attribution. All other rights reserved. 4763 Candidate biomarkers associated with GVHD severity at disease onset have been previously reported (reviewed in Chen and Zeiser7 and Adom et al8). Most of these biomarkers are cytokines or their soluble receptors, chemokines, growth factors, angiogenic factors, or molecules involved in inflammation. Some biomarkers are specifically associated with a GVHD target organ (skin, liver, or gut). Biomarker performances have been evaluated either separately, as composite panels, or as algorithms. Our group previously reported that fecal calprotectin was significantly associated with steroidrefractory gut GVHD and mortality9 with a C-index discrimination of 0.86 for a clinico-biologic score that included calprotectin, a-1 antitrypsin, and GVHD grade. However, American investigators at Ann Arbor have sequentially reported for more than a decade plasma biomarkers related to skin GVHD (Elafin or trappin10), gastrointestinal GVHD (regenerating islet-derived 3 a [REG3a]11,12), liver GVHD (hepatocyte growth factor [HGF]12), or different combinations of biomarkers.13-21 Specifically, these authors reported that 7 molecules could be potent GVHD biomarkers: Elafin, HGF, interleukin-2 receptor a (IL-2RA), IL-8 (CXCL8), REG3a, tumor necrosis factor a receptor-1 (TNFRI), and suppression of tumorigenicity 2 (ST2, the IL33 receptor). These 7 markers have been extensively studied in panels of 2, 3, 4, or 6 biomarkers or alone for ST2.14-19 Some of these panels have been internally or externally validated, usually using the same laboratory for biomarkers assays. External validations have been performed on different cohorts, but biomarkers assays have always been performed by authors who described seminal discoveries. Thus, our study is an external validation by independent investigators. Furthermore, it is worth noting that the time of measurement of these biomarkers varied significantly between these studies. Indeed, in what authors refers as panel 2 that includes ST2 and REG3a, which has been elegantly validated in a large international consortium, sampling was done 7 days after GVHD treatment initiation,14 whereas this same panel was also studied on samples taken 7 days after HSCT in another study.15 Panels including the 3, 4, and 6 biomarkers and ST2 have been tested on samples at GVHD diagnosis. All these panels are reported to predict GVHD prognosis and mortality. Finally, most recently, a randomized multicenter trial even used Ann Arbor biomarker status panel 2 to risk-stratify patients.22 We conducted a prospective single-center study to assess the prognostic value of plasmatic biomarkers using a single-center French population and local assays to analyze biomarkers. The 7 biomarkers Elafin, HGF, IL-2RA, IL-8, REG3, ST2, and TNFRI were measured at onset of GVHD, before corticostero (...truncated)