Comparative analysis of graft-versus-host disease prophylaxis with tacrolimus in combination with methylprednisolone or methotrexate after umbilical cord blood transplantation
INTERNATIONAL JOURNAL OF HEMATOLOGY
Authors: Shigemura, Tomonari; Sakashita, Kazuo; Okura, Eri; Morita, Daisuke; Komori, Kazutoshi; Kurata, Takashi; Hirabayashi, Koichi; Saito, Shoji; Tanaka, Miyuki; Yanagisawa, Ryu; Nakazawa, Yozo
Abstract
Post-transplant early immune disorders and engraftment failure/delay are major issues in unrelated umbilical cord blood transplantation (UCBT). We evaluated graft-versus-host disease (GVHD) prophylaxis approaches after UCBT by comparing UCBT outcomes with GVHD prophylaxis using tacrolimus plus methylprednisolone (Tac/mPSL, n = 32) to that with Tac plus methotrexate (Tac/MTX, n = 31) at a single pediatric transplantation center. The 30-day cumulative incidence rates of neutrophil engraftment and median neutrophil engraftment times in the Tac/mPSL and Tac/MTX groups were 70.1% and 90.3% and 19 and 17 days, respectively (p = 0.09). Pre-engraftment immune reactions (PIR) and acute GVHD were improved with Tac/MTX; PIR incidence (p = 0.020) and cumulative incidence of 100-day acute GVHD (grade II-IV, 38.7% vs 68.8%, p = 0.045; grade III-IV, 9.7% vs 34.4%, p = 0.021) were significantly lower in the Tac/MTX group than in the Tac/mPSL group. However, the incidence rates of relapse (p = 0.921) and cytomegalovirus reactivation (p = 0.908), and the estimated overall (p = 0.87) and event-free survival (p = 0.88) were comparable between the two groups. These data indicate that GVHD prophylaxis with Tac/MTX is associated with favorable results, including reduced PIR and acute GVHD incidence after UCBT, without adverse effects.
Symmetric Primitives with Structured Secrets
ADVANCES IN CRYPTOLOGY - CRYPTO 2019, PT 1
Authors: Alamati, Navid; Montgomery, Hart; Patranabis, Sikhar
Abstract
Securely managing encrypted data on an untrusted party is a challenging problem that has motivated the study of a wide variety of cryptographic primitives. A special class of such primitives allows an untrusted party to transform a ciphertext encrypted under one key to a ciphertext under another key, using some auxiliary information that does not leak the underlying data. Prominent examples of such primitives in the symmetric setting are key-homomorphic (weak) PRFs, updatable encryption, and proxy re-encryption. Although these primitives differ significantly in terms of their constructions and security requirements, they share two important properties: (a) they have secrets with structure or extra functionality, and (b) all known constructions of these primitives satisfying reasonably strong definitions of security are based on concrete public-key assumptions, e.g., DDH and LWE. This raises the question of whether these objects inherently belong to the world of public-key primitives, or they can potentially be built from simple symmetric-key objects such as pseudorandom functions. In this work, we show that the latter possibility is unlikely. More specifically, we show that: Any (bounded) key-homomorphic weak PRF with an abelian output group implies a (bounded) input-homomorphic weak PRF, which has recently been shown to imply not only public-key encryption but also a variety of primitives such as PIR, lossy TDFs, and even IBE. Any ciphertext-independent updatable encryption scheme that is forward and post-compromise secure implies PKE. Moreover, any symmetric-key proxy re-encryption scheme with reasonably strong security guarantees implies a forward and post-compromise secure ciphertext-independent updatable encryption, and hence PKE. In addition, we show that unbounded (or exact) key-homomorphic weak PRFs over abelian groups are impossible in the quantum world. In other words, over abelian groups, bounded key-homomorphism is the best that we can hope for in terms of post-quantum security. Our attack also works over other structured primitives with abelian groups and exact homomorphisms, including homomorphic one-way functions and input-homomorphic weak PRFs.