Severe Cytokine Release Syndrome Is a Fatal Complication after PBSC, but Not after BM Haploidentical Transplantation with Post-Transplant Cyclophosphamide

Abstract

Abstract Background: T-cell replete Haploidentical stem cell transplantation (Haplo-SCT) with high dose post-transplant cyclophosphamide (PT-Cy) represents an alternative option for patients with hematologic malignancies when a HLA identical sibling or a matched related donor is not available. Infusion-related febrile reaction and Cytokine release syndrome (CRS) were shown to occur within the first 24 hours and 14 days, respectively, after peripheral blood stem cell (PBSC) Haplo-SCT with PT-Cy. In particular, severe CRS is associated with an adverse outcome in terms of overall survival (OS) and non-relapse mortality (NRM) and appear to be mediated by a massive release of inflammatory cytokines, including IL-6, IFNγ, IL-2. Up to date, it is not known whether CRS is associated only with PBSC or may occur also after bone marrow graft. Moreover, no variable, either from the donor or the recipient, has been identified to be associated with the occurrence of CRS. Methods: We retrospectively analyzed 66 consecutive patients receiving Haplo-SCT with PT-Cy either form a BM (n=28) or PBSC (n=38) graft source between 2014 and 2015 at our institution. The two cohorts did not differ in terms of main patients' or donors' characteristics (median age of the recipient, pre-transplant disease status, CMV serostatus, HCT-CI, type of conditioning, donor age and type). CRS was identified and graded according to Abboud R, et al, BBMT 2016 and Lee DW, Blood 2014. Results: Grade 1 CRS characterized by febrile reaction was diagnosed in 34 out of 38 patients (89%) receiving PBSC and 20 out of 28 (71%) receiving BM Haplo-SCT with a median of 1 (0-7) and 2 (0-4) days, respectively, after transplant. By day 14 post-transplant CI of CRS grade ≥2 was 37% (95% CI: 19-45) after PBSC Haplo-SCT and 28% (95% CI: 8-52) after BM transplant (p=0.3). Interestingly, severe CRS grade ≥3 occurred only after PBSC transplant for a CI of 18% (95% CI: 3-44) vs 0% after BM Haplo-SCT (p=0.01) (Figure 1a). 1-year NRM did not differ between patients experiencing CRS ≥2 vs <2 either after BM (5%, 95% CI: 0-22 vs 12%, 95% CI: 0-44) or PBSC (37%, 95% CI: 12-63 vs 26%, 95% CI: 10-46; p=0.5) Haplo-SCT. Similarly, 3-year OS was not significantly different between subjects with CRS ≥2 vs <2: 52% (95% CI: 27-72) vs 65%(95% CI, 49-77) for the whole cohort (p=0.6). In marked contrast, recipients of PBSC Haplo-SCT experiencing grade ≥3 CRS had worse 1y-NRM compared with CRS <3: 78% (95% CI: 6-98) vs 20% (95% CI: 8-37) (p=0.002, Figure 1b). Consistently, 3-years OS after PBSC Haplo-SCT was significantly affected by CRS grade ≥3: 3-y OS was 21% (95% CI: 1-58) for CRS ≥3 vs 53% (95% CI: 25-75) for CRS <3 (p=0.01, Figure 1c). Within patients receiving PBSC Haplo-SCT, we were not able to identify any variable associated with the occurrence of severe CRS (data not shown). CRS ≥3 was the only variable associated with worse NRM both in univariate and multivariate analysis (HR: 4,7, 95% CI: 1,3-16,6, p=0.01). Pre-transplant disease status (CR vs PR vs PD/SD) and CRS ≥3 were associated with a worse 3-years OS both in univariate and multivariate analysis: disease status HR: 4,6, p=0.02; CRS≥3, HR: 13,4, 95% CI: 2,9-62,1; p=0.001. Conclusion: We have shown for the first time that CRS occurs not only after PBSC, but also BM Haplo-SCT with PT-CY. We have also found for the first time that severe CRS occurs only after PBSC transplant and confirmed that only CRS >3 is associated with a worse outcome. Deepening our understanding of CRS biology, by analyzing not only the T cell, but also the macrophage compartment, is warranted in order to improve our chances to predict and abrogate CRS after T-cell-replete Haplo-SCT. Figure 1. Cumulative incidence (CI) of CRS and disease outcome after Haplo-SCT. A) CI of CRS ≥3 for subjects receiving PBSC vs BM graft; b) 1-year NRM of CRS ≥ 3 vs <3 in recipients of PBSC Haplo-SCT; c) 3-year OS of CRS ≥ 3 vs <3 in recipients of PBSC Haplo-SCT Disclosures Carlo-Stella: ADC Therapeutics: Research Funding; Rhizen Pharmaceuticals: Research Funding; Boehringer Ingelheim: Consultancy, Honoraria. Santoro: Merck: Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees.