Pentraxin 3 As a Novel Diagnostic and Prognostic Biomarker for Acute GvHD and Fungal Infections in Adult Allogeneic HSCT Recipients

Abstract BACKGROUND: Pentraxin 3 (PTX3) is an essential component of the humoral arm of innate immunity, involved in controlling infections and modulating inflammation, and acting as a new prognostic and diagnostic biomarker in these conditions. PTX3 has been recently recognized as an extrinsic oncosuppressor in cancer; its role in hematological malignancies and allogeneic hematopoietic stem cell transplantation (allo-HSCT) is under investigation in both pediatric and adult settings. METHODS: From September 2012 we have conducted a prospective observational study to address the role of PTX3 in 35 adult patients affected by newly diagnosed AML (n=27) or ALL (8), and 73 adult pts who received allo-HSCT for acute leukemia. Stem cell donors were haploidentical (n=45), sibling (n=14), unrelated (n=14). Serial peripheral blood (PB) and bone marrow (BM) samples were collected from all patients. Time-points of sampling for leukemia patients included: diagnosis, relapse, any cycle of chemotherapy (CT; before starting and first day after the end), evaluation of treatment response (including BM), fever (first episode, or clinical suspicion of fungal infection). Samples from allo-HSCT patients were collected as following: before HSCT; at HSCT (day 0); day +7, day +14, 1 month (including BM), 2 months (including BM), 3 months (including BM) and 6 months (including BM) after HSCT; fever (first episode, or clinical suspicion of fungal infection); acute GvHD requiring systemic therapy (at onset or before starting a second line treatment, day+7 of treatment, then weekly during immunosuppressive therapy for 6 weeks). PTX3 level detection was performed with sandwich ELISA (detection limit 0.1 ng/ml). RESULTS: Firstly we analyzed the group of leukemia patients. Although PTX3 levels on PB at leukemia diagnosis were not influenced by patients (age, sex) or disease characteristics (type of leukemia, BM and PB blasts), they correlated with white blood cell counts (p=0,03). No statistical difference was found matching PTX3 levels at diagnosis with disease relapse or overall survival (OS). We did not observe any correlation between PTX3 levels before starting CT and response to treatment. When we moved to examine allo-HSCT patients, we observed an important difference in PTX3 levels on PB at day 0 among patients treated with treosulfan or busulfan-based conditioning (p=0,027). We found a trend towards an increase of PTX3 levels for patients who developed a subsequent GvHD at day +7, and for patients who received ATG instead of PT-Cy at day +14 (p =0,06 for both of them). Moreover, we evaluated 38 events of acute GvHD. Interestingly, PTX3 levels on PB at GvHD onset predict a subsequent evolution to chronic GvHD (p=0,02), also confirmed by ROC analysis (AUC 0.786) which allowed us to identify a threshold of 58 ng/ml (sens 75%, spec 85%). At day +7 from the start of therapy, PTX3 levels were correlated directly with steroid-refractory GvHD (p=0,004) and reflare (p=0,024), and inversely with GvHD resolution (p=0,003). Concomitantly, the ROC analysis identified a threshold of 18 ng/ml as predictor for steroid-resistance (AUC 0,91; sens 75%, spec 85%) and GvHD resolution (AUC 0,848; sens 70%, spec 85%). Finally, we reported 142 infectious events. No difference in PTX3 levels was reported for first and subsequent episodes, or with bloodstream infections. Importantly, in presence of invasive fungal infection (IFI), the levels of PTX3 were significantly increased (p=0,016). CONCLUSIONS: PTX3 is an early biomarker for major transplant-related complications, as IFI and acute GvHD. This study suggests that PTX3 is able to predict GvHD resolution and the risk of subsequent chronic GVHD development; moreover PTX3 predicts steroid-refractory GvHD, offering a new tool in pindividualized clinical management. Disclosures Ciceri: MolMed SpA: Consultancy.