Work

Incorporation of thiotepa in a reduced intensity conditioning regimen may improve engraftment after transplant for HLH

Year: 2020

Type: article

Abstract: Haematopoietic stem cell transplantation (HSCT) for patients with haemophagocytic lymphohistiocytosis (HLH) following myeloablative conditioning regimens (MAC) is associated with high rates of non-relapse mortality (Baker et al., 2008). Reduced intensity conditioning (RIC) strategies have shown improved survival (Shenoy et al., 2005; Cooper et al., 2008) but are associated with high rates of mixed chimerism and graft failure particularly with the use of conditioning regimens incorporating alemtuzumab (Marsh et al., 2013). Allen et al. describe outcomes after RIC strategy for HLH and other primary immune deficiency syndromes in a prospective, phase 2 multi-centre (RICHI) trial implemented by the Blood and Marrow Transplant Clinical Trials Network using fludarabine, melphalan and alemtuzumab (day-14). One-year overall survival estimate for patients with HLH (n = 34) was 82.4% (90% CI: 68.4, 90.6%) and 18-month survival was 68% (90% CI: 51.8–79.7). As with other alemtuzumab-based RIC approaches, sustained engraftment was poor with fewer than half (41% of patients with HLH) being engrafted without secondary graft failure, need for donor lymphocyte infusion (DLI) or second transplant and alive at one year. Although the survival outcomes compared favourably to MAC regimens, long-term durable engraftment was suboptimal (Allen et al., 2018). We report our results incorporating thiotepa into the fludarabine/melphalan/alemtuzumab (day-14) backbone (Marsh et al., 2013; Allen et al., 2018). In recent reports, the addition of thiotepa during conditioning was shown to be safe and was associated with enhanced engraftment across a range of donor sources (Ciurea et al., 2012; Larsen et al., 2013). In the cohort described here, a single dose of 10 mg/kg thiotepa was added to the fludarabine/melphalan/alemtuzumab regimen on day –3 (RICHI+TT). Graft-versus-host disease (GVHD) prophylaxis was similar to the published regimen using methylprednisolone through day +28 (followed by wean as tolerated) and cyclosporine through day 180. This study is a retrospective analysis of nine consecutive patients treated with RICHI+TT approved by the Baylor College of Medicine Institutional Review Board. Patient characteristics and outcomes are summarized in Table 1. Our cohort consisted of eight males and one female with a median age of seven years (range 1–18 years). Seven patients had HLH with proven pathogenic genetic mutations [biallelic PRF1 – 2, UNC13D – 2, STXBP1 – 1, RAB27A –1, STAT3 gain of function (GOF) – 1], while the other two patients had HLH without identified pathogenic mutations [chronic active Epstein–Barr virus (CAEBV) – 1, juvenile idiopathic arthritis with refractory macrophage activation syndrome (MAS) – 1]. STAT3 gain-of-function mutation (c.857A>G (p.E286G)). Additionally, heterozygous mutation STXBP2 (c.1034T>C) and a homozygous NLRP4 mutation (c.281G>C (p.G94A)) Seven patients received treatment with dexamethasone and etoposide as per the HLH-94 or HLH-2004 protocols. Of these seven patients, two received additional cytokine-directed therapy, and one of these patients also received alemtuzumab as HLH salvage therapy. One patient received dexamethasone monotherapy. One patient with a congenital homozygous PRF1 mutation was diagnosed based on screening performed due to family history and had no clinical indication for HLH-directed therapy prior to transplant. Six patients received stem cell grafts from a matched related donor, two from a mismatched unrelated donor, and one from a matched unrelated donor. Eight patients received a bone marrow product and one received a peripheral blood stem cell product. All patients engrafted at a median of 15 days (range: 11–21 days) post-transplant. At initial testing for donor engraftment, eight patients were 100% donor and one patient had 99% donor chimerism. We compared our long-term outcomes using the same parameters for sustained donor engraftment defined in the RICHI study as >5% donor chimerism without any intervention and alive at one year post-transplant. Six of the nine patients were evaluable to assess donor chimerism at one year with a median follow-up of 875 days (range: 366–1000 days) and all had >5% donor chimerism. Five of the six evaluable patients met criteria for sustained donor engraftment, and all maintained 100% donor chimerism at last follow-up without need for any intervention (Table 1). Only one of the six (17%) patients had evidence of falling donor chimerism which stabilized at 40% donor chimerism after DLI. No other patient required use of DLI or a second transplant. No patient had primary or secondary graft failure. These results compare favourably with the cohort described by Allen et al. where only 14 out of 28 (50%) evaluable patients with HLH achieved or maintained sustained engraftment without intervention using DLI or a second transplant at one year (Table 2). Three patients in our cohort were not evaluable for long-term chimerism assessment due to death prior to one year. One patient with an underlying STAT3 GOF mutation developed a severe HLH flare two months post-HSCT which could not be controlled with immunosuppressive therapy despite 90% donor chimerism at the time. The second patient had HLH associated with severe systemic juvenile idiopathic arthritis (JIA). He engrafted early at day +10 with 100% donor chimerism but developed fulminant engraftment syndrome with rapidly progressive multi-organ failure and died on day +12. The third patient had underlying CAEBV and developed Grade IV GVHD six months post-transplant and died from complications of intensive immunosuppression. Of note, all three patients that died had HLH associated with conditions that have been associated with poor outcomes after HSCT, namely STAT3 GOF, systemic JIA and CAEBV (Snowden et al., 2012; Snowden et al., 2017; Sawada & Inoue, 2018; Fabre et al., 2019). Three patients developed GVHD: one patient had Grade IV gut GVHD as described above and two patients developed Grade II GVHD which resolved with therapy. No patient developed toxicities attributable to thiotepa. While viral reactivations were frequent, no patient developed viral disease. There were no cases of invasive fungal infections. Six of the nine patients described here are alive and disease-free with stable long-term engraftment. In this cohort, thiotepa was added to the conditioning backbone of fludarabine, melphalan and alemtuzumab and was well tolerated. While this was a retrospective series and not directly comparable to the RICHI study, the RICHI+TT approach achieved relatively favourable long-term donor engraftment with a decreased need for DLI and second transplants. Other strategies to improve engraftment may include incorporation of alternative chemotherapy such as treosulfan (Messina et al., 2018), use of immunotherapies such as CD45 antibody or immunotoxin (Palchaudhuri et al., 2016), and/or use of personalized approaches with targeted alemtuzumab-based pharmacokinetics. Additionally, treatment with emapalumab (neutralizing IFN-γ) prior to HSCT may be associated with improved engraftment (Jordan et al., 2019). Further prospective studies are needed to determine the safest and most effective treatment strategies for patients with HLH who require HSCT for cure. SN, OE, GS, CAM performed the research. CAM, CAE, RAK designed the research study. SN, CAM analyzed the data. SN, HEH, RAK, CAE, CAM wrote the paper. HEH is a founder with equity in Allovir and Marker Therapeutics, receives research support from Tessa Therapeutics and Cell Medica and has served on advisory boards for Kiadis, Gilead Biosciences and Novartis. CAE serves on the Novimmune Scientific Steering Committee for emapalumab clinical trials (unpaid).

Source: British Journal of Haematology

DOI: 10.1111/bjh.16370