C\ Taylor & Francis
`
`~
`
`Taylor&Fr•ncisGroup
`
`rom
`OPINION
`
`Expert Opinion on Orphan Drugs
`
`ISSN: (Print) 2167-8707 (Online) Journal homepage: https://www.tandfonline.com/loi/ieod20
`
`Therapeutic targets and emerging treatment
`options in gastrointestinal acute graft-versus-host
`disease
`
`Anne S. Renteria, John E. Levine &James L. M. Ferrara
`
`To cite this article: Anne S. Renteria, John E. Levine & James L. M. Ferrara (2016) Therapeutic
`targets and emerging treatment options in gastrointestinal acute graft-versus-host disease , Expert
`Opinion on Orphan Drugs, 4:5, 469-484, DOI: 10.1517/21678707.2016.1166949
`
`To link to this article: https://doi.org/10.1517 /21678707.2016.1166949
`
`Accepted author version posted on line: 29
`Mar 2016.
`Published online: 06 Apr 2016.
`
`Submit your article to this journal 8"
`
`Wtl. Article views: 196
`~ ... ~ View related articles 8"
`
`(I} View Crossmark data 8"
`
`CronM.irk
`
`~ Citing articles: 1 View citing articles 8"
`
`Full Terms & Conditions of access and use can be found at
`https://www.tandfonline.com/action/journallnformation?journa1Code=ieod20
`
`Pharmacvclics Exhibit 2049
`Sandoz v. Pharmacyclics
`IPR20 19-00865
`
`

`

`EXPERT OPINION ON ORPHAN DRUGS, 2016
`VOL. 4, NO. 5, 469- 484
`http://dx.doi.org/1 O. l 517/21678707.2016. l 166949
`
`REVIEW
`
`W Taylor & Francis
`~ Taytor &.Francis Group
`
`Therapeutic targets and emerging treatment options in gastrointestinal acute
`graft-versus-host disease
`Anne S. Renteria•, John E. Levine" and James L. M. Ferrarab
`
`•Blood and Marrow Transplantation Program, Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA; bHematologic
`Malignancies Translational Research Center, Blood and Marrow Transplantation Program, Tisch Cancer Institute, Icahn School of Medicine at Mount
`Sinai, New York, NY, USA
`
`ABSTRACT
`Introduction: Graft-versus-host disease (GVHD) continues to be the major lethal complication of
`allogeneic hematopoietic stem cell transplantation (HCT) but the standard of care, high dose steroids,
`has not changed in 40 years. Approximately 50% of GVHD patients will develop steroid refractory
`disease, typically involving the gastrointestinal (GI) tract, which has a very poor prognosis. Newly
`developed GVHD biomarker-based risk scores provide the first opportunity to treat patients at the
`onset of symptoms according to risk of steroid failure. Furthermore, improvements in our understand(cid:173)
`ing of the pathobiology of GVHD, its different signaling pathways, involved cytokines, and the role of
`post-translational and epigenetic modifications, has identified new therapeutic targets for clinical trials.
`Areas covered: This manuscript summarizes the pathophysiology, diagnosis, staging, current and new
`targeted therapies for GVHD, with an emphasis on GI GVHD. A literature search on PubMed was
`undertaken and the most relevant references included.
`Expert Opinion: The standard treatment for GVHD, high dose steroids, offers less than optimal out(cid:173)
`comes as well as significant toxicities. Better treatments, especially for GI GVHD, are needed to reduce
`non-relapse mortality after allogeneic HCT. The identification of high risk patients through a biomarker(cid:173)
`defined scoring system offers a personalized approach to a disease that still requires significant research
`attention.
`
`ARTICLE HISTORY
`Received 13 January 2016
`Accepted 14 March 2016
`Published online
`5 April 2016
`
`KEYWORDS
`Biomarkers; GVHD;
`graft-versus-host disease;
`microbiome; Paneth cells;
`steroid-refractory GVHD;
`treatment
`
`1. Introduction
`
`Allogeneic hematopoietic stem cell transplantation (HCTI is
`increasingly used to cure malignant and benign hematologic
`diseases, with over 8000 transplants performed in the year
`2013.[1] Transplanted T cells from the donor can recognize
`and eradicate hematologic malignancies through the immu(cid:173)
`nologic graft-versus-leukemia
`(GVL) effect. Unfortunately,
`donor conventional T cells (Tcons) recognize normal recipient
`tissues and attack them, causing graft-versus-host disease
`(GVHD). The skin, liver, and gastrointestinal (GI) tracts are the
`primary targets of acute GVHD, which is the major cause of
`nonrelapse mortality (NRM) after HCT, [2] and develops in 40-
`60% of patients.[3]
`GVHD in the skin is the most frequently involved target
`organ and presents as an erythematous maculopapular rash;
`liver GVHD, the least common, causes hyperbilirubinemia; GI
`GVHD involves the upper GI tract, causing nausea, vomiting,
`and anorexia and, more often, the lower GI tract, cau sing
`diarrhea and abdominal pain. Each GVHD target organ is
`staged on a 0-4 severity scale, and the individual stages are
`used to create a composite clinical severity grade (Table 1).
`[4-6] Onset clinical severity does not correlate as well with
`survival as does maximal clinical severity, which also reflects
`treatment response. Patients with significant (;:,:Grade 2)
`
`GVHD are all treated similarly with high-dose steroid s, with
`intensification reserved for primary treatment failure. The
`high rate of treatment failure for lower GI GVHD accounts
`for the majority of NRM in the first 6 months after HCT.[7,8]
`Therefore, this manuscript will emphasize GI GVHD and the
`available and emerging therapies for its treatment.
`
`2. Pathophysiology
`
`The graft-versus-host (GVH) reaction is initiated when donor
`Tcons
`respond
`to genetically defined protein antigens
`expressed on host antigen-presenting cells (APCs).[2] Donor
`Tcons proliferate and differentiate during GVH, and the balance
`between effector and regulatory T cells (Tregs) plays an impor(cid:173)
`tant role in its progression and resolution. This paradigm for
`GVHD pathophysiology involves three distinct phases. In the
`first phase, which commences weeks before the onset of symp(cid:173)
`toms, tissue damage from the radiation and/ or chemotherapy
`given in the conditioning regimen initiates an inflammatory
`immunologic cascade involving both the innate and adaptive
`immune systems. The release of proinflammatory cytokines (e.g.
`TNFa, IL-1, and IL-6) promotes the activation of host APCs,
`which in turn drive donor Tcon proliferation, differentiation,
`and migration to target tissues.[1 OJ Damage to the GI epithe(cid:173)
`lium allows the translocation from the gut lumen of danger
`
`CONTACT Anne S. Renteria Q Anne.Renteria@mountsinai.org G) Blood and Marrow Transplantation Program, Tisch Cancer Institute, Icahn School of Medicine at
`Mount Sinai, New York, NY 10029, USA
`© 2016 lnforma UK Limited, trading as Taylor & Francis Group
`
`

`

`470 @ A. S. RENTERIA ET AL.
`
`Article highlights
`
`• Acute GVHD develops in 40-60% of allogeneic HCT recipients and is
`the major cause of NRM. The standard first-line therapy for acute
`GVHD is high-dose steroids, but SO% of cases are steroid refractory.
`GI GVHD accounts for the majority of NRM because of its high rate of
`treatment failure.
`• New treatments for GVHD under study that exploit new t argets
`involved in GVHD pathophysiology include blockade of leukocyte
`trafficking, JAK inhibition, hi stone deacetylase inhibitors, alpha-1
`antitrypsin, induction of regulatory T cells, and restoration of GI
`barrier function through cytokines.
`• A va lidated scoring system based on GVHD biomarkers (Ann Arbor
`risk scores) objectively stratifies patients according to risk of primary
`treatment failure and can identify patients for clinical trials before
`steroid refractory GVHD ha s developed.
`• The loss of diversity in the GI microbiota is associated with increased
`GVHD mortality and relates to different factors, including exposure t o
`antibiotics. Preservation or restoration of a healthy GI microbiome is
`an alternative strategy to treat GI GVHD.
`
`This box summarizes key points contained in the article.
`
`signals and pathogen-associated molecular patterns (PAMPs)
`such as bacterial cell wall components (e.g. lipopolysaccharide
`- LPS) and damage-associated molecular patterns (DAMPs) (e.g.
`ATP and extracellular matrix proteins) to amplify the cytokine
`cascade and trigger the development of chemokine gradients
`that attract donor Tcons.[11 ] Tregs are transcription factor fork(cid:173)
`head box P3 (Foxp3+) CD4 + T cells that by suppressing allor(cid:173)
`eactive lymphocytes can dampen the effect of GVHD caused by
`Tcons [12] (Figure 1 ). In the second phase, also days to weeks
`before the onset of symptoms, T-cell traffic to target organs in a
`highly regulated process, during which interactions between
`adhesion molecules (e.g. MAdCAM-1) and integrins (e.g. a4~7)
`result in leukocyte adherence to the capillary endothelium and
`migration into the subendothelium [13] as shown in Figure 2.
`Chemokines regulate not only the trafficking of leukocytes, but
`also their activation and differentiation by binding to specific
`receptors, such as chemokine (C-C motif) receptor 5 (CCRS),
`CCR6, and CCR7.[l 4, 15] In Phase 3, clinical symptoms com(cid:173)
`mence when Tcons cause tissue destruction through direct
`cytotoxic activity, mostly through Fas ligand: Fas and perforin(cid:173)
`granzyme pathways [16], as well as through cytokines such as
`TNFa (Figure 3).
`
`3. Diagnosis
`
`The diagnosis of acute GI GVHD is based on clinical signs and
`symptoms
`(which do not occur until Phase 3 of its
`pathophysiology).
`It is common practice to rule out infectious enteritis by
`testing the stool for adenovirus, rotavirus, or Clostridium diffi(cid:173)
`cile.[17] Diarrhea
`from pretransplant chemoradiotherapy(cid:173)
`induced GI damage is common prior to engraftment and
`when GVHD symptoms occur; most centers perform diagnos(cid:173)
`t ic endoscopy to establish the diagnosis. However, endoscopic
`abnormalities are seen in less than one-third of GI GVHD cases
`and are nonspecific [18]; biopsies are usually performed to
`obtain histologic confirmation of the diagnosis. Flexi ble sig(cid:173)
`moidoscopy is as sensitive as full colonoscopy.[19] Upper GI
`endoscopy is often performed, as the small bowel may be the
`major source of diarrhea, although this approach is not uni(cid:173)
`versally accepted.[20] Although GI GVHD symptoms may start
`as early as 9 days after HCT, they often do not start until
`patients have been discharged from the hospital, and accurate
`measurements of outpatient diarrheal output are not routinely
`available. The average volume per episode of diarrhea in an
`adult has been estimated to be - 200 ml, allowing for staging
`and grading of GI GVHD when only the number of diarrhea
`episodes is known.[9]
`
`4. Histology
`
`GI crypt dropout is characteristic of histologic GVHD, and
`intestinal stem cells (ISCs) are considered targets of the dis(cid:173)
`ease,[2 1,22] with crypt cell apoptosis a histologic hallmark.
`Nonetheless, GI GVHD histology is often not straightforward.
`Crypt damage from conditioning chemotherapy ± radiation
`can take weeks to heal, and persistent changes can overlap
`with the onset of GI GVHD symptoms. Other causes of crypt
`damage include infections and mycophenolate mofetil,[23] a
`commonly used immunosuppressant. The patchy nature of GI
`GVHD histology sometimes leads to false-negative biopsies
`through sampling error. Finally, the histologic grading system
`(Table 2) [24] has never been standardized for a number of key
`parameters, including the number of tissue sections or high(cid:173)
`powered fields to be analyzed, or the number of apoptotic
`
`Table 1. GVHD target organ staging.
`
`Stage
`0
`
`Skin (active erythema only)
`No active (erythematous) GVHD rash
`
`Liver (bilirubin)
`
`Upper GI
`
`<2 mg/ di
`
`2-3 mg/d i
`
`No or intermittent nausea,
`vomiting, or anorexia
`Persistent nausea,
`vom iting, or anorexia
`
`2
`
`4
`
`3.1 - 6 mg/ di
`
`6.1-15 mg/ di
`
`>15 mg/di
`
`Maculopapular ra sh
`< 25% BSA
`Maculopapular ra sh
`25-50% BSA
`Maculopapular rash
`>50% BSA
`Generalized erythroderma (>50% BSA) plus
`bullous formation and desquamation >5% BSA
`*When stool volume is not quantified, a 200-ml/episode can be used as an estimate for adults [9].
`Overall cli nical grade (based on most severe target organ involvement):
`Grade 0: No stage 1-4 of any organ.
`Grade I: Stage 1-2 skin without liver, upper GI, or lower GI involvement.
`Grade II: Stage 3 rash and/o r Stage 1 liver and/or Stage 1 upper GI and/or Stage 1 lower GI.
`Grade Ill: Stage 2-3 liver and/or Stage 2- 3 lower GI, with Stage 0-3 skin and/or Stage 0-1 upper GI.
`Grade IV: Stage 4 skin, liver, or lower GI involvement, with Stage 0-1 upper GI.
`
`Lower GI (stool output/day)*
`Adult: <500 ml/ day or <3 episodes/ day
`Child: < 10 ml/ kg/ day or <4 episodes/ day
`Adult: 500- 999 ml/ day or 3-4 episodes/day
`Child: 10- 19.9 ml/ kg/ day or 4-6 episodes/ day
`Adult: 1000- 1500 ml/ day or 5-7 episodes/day
`Child: 20- 30 ml/ kg/day or 7-1 0 episodes/ day
`Adult: > 1500 ml/ day or > 7 episodes/ day
`Child: > 30 ml/kg/ day or > 10 episodes/day
`Severe abdominal pain with or without ileus or
`grossly bloody stool (regardless of stool vo lume)
`
`

`

`EXPERT OPINION ON ORPHAN DRUGS @ 471
`
`Donor Tcon Activation
`
`Figure 1. Blockade of donor Tcon activation (first phase of acute GVHD).
`This schematic figure depicts key events in early acute GVHD pathophysio logy in the GI tract, the most important GVHD target organ. A villus of the small intestine
`(left) is highly magnified (right).
`The tran splant conditioning regimen leads to generation of danger signals and pro-inflammatory cytokines (e.g. TNFa, IL-6) that activate APCs. For illustration
`purposes, one APC is shown interacting with one conventional T cell (Tcon) and one regulatory regulatory T cell (Treg). Physiologically, these cells are clustered
`beneath crypts, in the lamina propria. Etanercept and infliximab, two anti-TNFa antibodies, neutralize TNFa directly. Tocilizumab, prevents APC activation by
`inhibiting binding of IL-6 to the IL-6 receptor.
`Steroids, ATG, a/emtuzumob, and MSCs are broad immunosuppressors that act on multiple immunologic processes, including the activation and differentiation of
`Tcons through different mechanisms (see text). ECP and vorinostat suppress host APC and promote Treg expansion.
`Abbreviations: APC, antigen-presenting cell; ATG, antithymocyte globulin; ECP, extracorporeal photopheresis; MSC, mesenchymal stromal cell.
`
`cells required to diagnose Grade 1 GVHD. As a consequence,
`interobserver variability is high, histologic severity does not
`correlate well with clinical symptoms, and a negative biopsy
`does not necessarily rule out GVHD.[25] Paneth cells are pri(cid:173)
`marily located in the crypts of the small intestine, and their
`number inversely correlates with high-risk disease.[26] Paneth
`cells are readily identified by their location in the GI tract and
`their histochemical staining with lysozyme. Their straightfor(cid:173)
`ward quantification can aid in establishing the diagnosis and
`in prognosticating its severity.
`
`antimicrobial peptide and regulator of intestinal gram-positive
`bacteria. As already mentioned, Paneth cell loss correlates with
`GI GVHD severity and long-term outcomes.[26] As a biomarker
`specific for GI GVHD, REG3a discriminates between GVHD and
`non-GVHD causes of diarrhea.[32] ST2 is secreted in response
`to inflammatory stimuli, and it functions as a decoy receptor
`for interleukin-33 (IL-33), which drives Tcons toward a proin(cid:173)
`flammatory phenotype.[33] ST2 plasma concentrations at the
`initiation of GVHD treatment strongly correlates with eventual
`resistance to treatment and 6-month NRM, which is primarily
`driven by steroid-refractory (SR) GI GVHD.[34]
`
`5. GVHD biomarkers
`
`Despite a large number of plasma proteins, DNA single
`nucleotide polymorphisms, microRNA molecules, and periph(cid:173)
`eral blood cellular subsets with associations to GVHD,[27-29]
`only a small number of candidate biomarkers have been
`validated in multicenter patient cohorts. Of these validated
`biomarkers, those with greatest relevance to GI GVHD are
`tumor necrosis factor receptor-1 (TNFRl ), regenerating islet(cid:173)
`derived protein-3-a (REG3a}, and suppression of tumorigeni(cid:173)
`city 2 (ST2). TNFRl, a membrane receptor for TNF-a that
`becomes soluble after binding its ligand, is not specific for GI
`GVHD but has been shown to strongly correlate with overall
`GVHD severity, response to treatment, NRM, and survival.
`[30,31] REG3a, which
`is secreted by Paneth cells,
`is an
`
`5. 1. Biomarker-defined risk stratification
`
`Individual GVHD biomarker concentrations vary widely among
`centers. To overcome this limitation, the Ann Arbor (AA) scor(cid:173)
`ing system used several biomarkers from a multicenter cohort.
`An algorithm combining the plasma concentrations of TNFRl,
`REG3a, and ST2 at GVHD onset was developed to categorize
`patients according to risk of primary treatment failure and
`NRM.[35] Thresholds define three distinct scores: AA 1, NRM
`-1 0%; AA2, NRM -25%; and AA3, NRM -40%. Because relapse
`rates do not differ among the AA scores, these differences in
`NRM translate into significant differences in overall survival. It
`is important to note that AA scores identify patients who will
`later develop lower GI GVHD but who present with only a rash.
`
`

`

`472 @ A. S. RENTERIA ET AL.
`
`lnolimomab
`Denlleukin diftitox
`
`Nucleus:
`HDACi
`
`Cytoplasm:
`SYKi
`JAKi
`AAT
`
`Tcon Proliferation & Migration
`
`Figure 2. Blockade of Tcon proliferation and migration (second phase of acute GVHD).
`Following activation by host APCs, donor Tcons proliferate and migrate from secondary lymphoid organs and tissues associated with the mucosa (e.g. Peyer's
`patches). Activated Tcons release cytokines (e.g. IL-2) that promote further proliferation and differentiation. Basiliximab, daclizumab, inolimomab, and denileukin
`diftitox all bind to IL-2 receptor (IL-2R). HiDAC inhibitors suppress APCs activity, enhance Treg activity, and reduce pro-inflammatory cytokines through DNA
`methylation (azacitidine) and histone acetylation (vorinostat), whi le inhibitors of JAK1/2 (ruxolitinib), SYK (fostamatinib, and a-1-antitrypsin (AAD suppress cytokine
`production through cytoplasmic receptors. The migration of activated Tcon s (T cel l trafficking) into the GI subendothelium requires interaction between integrins
`(e.g. a4P7-integrins) expressed on Tcons and their receptors. Natalizumab and vedolizumab are antibodies that block a4 or a4P7, respectively. Intestinal stem cells
`(ISCs) are located at the base of th e crypts, interspersed with Paneth cells, and are responsible for crypt regeneration. Type 3 innate lym phoid cells (I LC3 s) produce
`IL-22 which is trophic for ISCs and which induces regenerating islet derived protein-3-a (REG3a), an antimicrobial peptide which destroys gram-positive bacteria.
`Abbreviations: APC, antigen-presenting cell; GI, gastrointestinal; GVHD, graft-versus-host disease.
`
`Patients whose skin was the only target organ affected at the
`time of diagnosis and who were classified as AA3 by biomar(cid:173)
`kers were twice as likely to develop lower GI GVHD later as
`patients classified as AA 1.[35] Biomarker-defined GVHD sever(cid:173)
`ity thus appears promising as an opportunity for early inten(cid:173)
`sive intervention in patients diagnosed with high-risk GVHD.
`
`6. Therapeutic options
`
`6. 1. Broad immunosuppression
`
`Most therapeutic options for GVHD have been evaluated in
`the context of systemic disease where sometimes, only overall
`grades are reported. In this review, we highlight outcomes for
`GI GVHD wherever possible; otherwise, the results described
`apply to GVHD in general.
`Intensified immunosuppression with systemic steroids is
`the only proven treatment for GVHD. Unfortunately, SR
`GVHD develops in - 50% of patients,[7,36,37] more often in
`patients with lower GI involvement,[?] and there is no estab(cid:173)
`lished second-line therapy. The response rates to second-line
`treatment are typically low (20-40%), and survival is poor,
`highlighting the urgent need
`for better therapies.[36,38]
`Thus, SR GVHD remains a significant contributor to treat(cid:173)
`ment-related mortality (TRM) and morbidity. As reviewed in
`the following section and summarized in Tables 3 and 4,
`
`several treatments have shown benefit in single-arm studies;
`however, that benefit has not been confirmed in the few that
`have been evaluated by more rigorous randomized trials.
`
`6.2. Steroids
`
`GVHD treatment usually begins with high doses of prednisone
`or, for patients unable to take oral steroids, methylpredniso(cid:173)
`lone (MP) (Figure 1 ). A randomized, prospective clinical trial
`showed that a 50% reduction in the starting dose of steroids
`was effective, although patients with more severe GVHD were
`more likely to require addition of secondary immunosuppres(cid:173)
`sion.[39] Upper GI GVHD may also respond well to the combi(cid:173)
`nation of 1 mg/ kg/d of prednisone and nonabsorbable steroid
`therapy.[40] Conversely, doses higher than 2 mg/ kg/day do
`not improve outcomes.[41]
`
`6.2. 1. Nonabsorbable steroids
`Oral nonabsorbable steroids, such as budesonide and beclo(cid:173)
`methasone, theoretically deliver high steroid doses to the GI
`tract without incurring the side effects of systemic steroid
`therapy. These agents have primarily been tested in combina(cid:173)
`tion with systemic steroids as part of first-line therapy for GI
`GVHD. It should be noted, however, that the term 'nonabsorb(cid:173)
`able' is a misnomer, because synthetic steroid screens often
`demonstrate significant absorption.[42] Results of
`this
`
`

`

`EXPERT OPINION ON ORPHAN DRUGS @ 473
`
`I
`"
`
`\
`
`' /
`Enter~ occus -' aJ 11u,
`
`I
`
`;
`
`'
`
`....
`Target Cell
`Destruction
`
`• Reg3a•enters
`bloodstream
`
`!£/--JU/
`
`o:01sMountSmaf
`IJN1th.5yllm
`
`Figure 3. Prevention of target cell destruction (third phase of acute GVHD).
`Activated Tcons that migrated to the GI epithelium secrete inflammatory cytokines and lyse ISCs and Paneth cells, disrupting the GI mucosal barrier. REG3a stored
`intracellularly and in the mucus leaks into the systemic circulation along with luminal contents, such as bacteria. Strategies to restore the integrity of the mucosal
`barrier are currently explored via clinical trial of recombinant IL-22 (r/L-22) administration.
`Abbreviations: GI, gastrointestinal; GVHD, graft-versus-host disease.
`
`Table 2. Histologic grading for acute gastrointestinal graft-versus-host disease
`(24].
`
`Grade
`0
`1
`2
`3
`4
`
`Histology
`
`Normal mucosa
`Isolated apoptotic epithelial cells without crypt loss
`Loss of isolated crypts without loss of contiguous crypts
`Loss of two or more contiguous crypts
`Extensive crypt loss with mucosal denudation
`
`Marrow Transplant Clinical Trials Network. As a salvage ther(cid:173)
`apy for SR GVHD, response rates of 57-76% have been
`reported for sirolimus.[48,53] Its side effect profile differs
`from other immunosuppressants and
`includes transplant(cid:173)
`related microangiopathy (TMA), particularly when combined
`with calcineurin inhibitors, as well as hyperlipidemia.[53]
`
`approach have been mixed, [43,44] and the only Phase Ill trial
`involving budesonide did not meet its primary end point of
`time to treatment failure.[40] Steroid delivery through arterio(cid:173)
`graphy to the arterial blood supply to the gut has been
`explored in several small studies. In two prospective series,
`GI GVHD response rates of 70-85% were observed and
`appeared better than historical controls.[45-47] The need for
`trained and available interventional radiology specialists may
`explain the lack of larger and adequately powered rando(cid:173)
`mized studies.
`
`6.3. Sirolimus
`
`Sirolimus, an mTOR inhibitor, blocks late Gl cell cycle progres(cid:173)
`sion, most prominently in T lymphocytes,[48] and also main(cid:173)
`tains Treg populations.[49] Sirolimus has proven efficacy in
`GVHD prophylaxis [50,51 ] and showed comparable activity to
`that of high-dose steroids as primary therapy for acute GVHD
`in one study.[52] Sirolimus is currently being compared to
`prednisone as a single agent for primary therapy of low-risk
`GVHD in a randomized trial conducted by the Blood and
`
`6.4. Purine synthesis inhibitors
`
`Mycophenolate mofetil (MMF) blocks de nova purine synthesis
`through inhibition of inosine monophosphate dehydrogenase
`(IMPDH), which arrests
`lymphocytes,
`in S-phase.[33]
`Encouraging results of MMF as primary treatment for GVHD
`were observed in several small series, [54,55] including multi(cid:173)
`center, randomized, Phase II trial.[56] However, a randomized,
`placebo-controlled trial was closed to accrual early for futility.
`[57] A significant drawback to MMF is its GI toxicity profile,
`which includes ulcerative esophagitis, reactive gastropathy,
`and pathologic changes very similar to GVHD that can cloud
`diagnostic biopsy interpretation.[58]
`Pentostatin, a nucleoside analog, inhibits adenosine deami(cid:173)
`nase and induces cell death particularly in T and NK cells.[59]
`Pentostatin has shown mixed results for SR GVHD in small
`series, with survival ranging from 7 to 43%.[60- 62]
`
`6.5. Anti-T-ce/1 serotherapy
`
`Antithymocyte globulin (ATG) consists of polyclonal lgG anti(cid:173)
`bodies against human T cells (Figure 1 ). The role of both
`
`

`

`)>
`----l
`m
`►
`::0
`m
`----l
`m z
`::0
`~
`?>
`®
`... .._, ...
`
`'
`
`(Continued)
`
`were used. High rate of infections
`
`Limited activity. Variety of doses and schedules
`
`ORR 38% (CR 21%); 2 year Schmitt (2011) [62]
`ORR 50% (CR 33%)
`Pida/a (2010) [61]
`
`Alam (2013) [60]
`
`additional therapies
`
`ORR 33%; 53% received
`
`05 17%
`
`(2015) [55]
`
`Inagaki
`
`[57]
`
`Bolanos-Meade (2014)
`
`3 year OS 86%
`day 56 CR 100% (GI only);
`Day 28 ORR 83% (GI only);
`
`from MMF
`(p = 0.08); no benefit
`1 year relapse 24 vs. 16%
`(p = 0.34);
`1 year OS 58% vs. 65%
`(p = 0.34);
`
`Day 56 CR 60 vs. 54%
`
`16 vs. 52%
`pts for toxicity; 1 year OS
`MMF discontinued in 21%
`31%);
`
`responses
`
`survival
`
`Furlong (2009) [54]
`
`ORR 47 vs. 48% (CR 31 vs.
`
`Day 35 response
`
`ORR 62% (CR 46%); 1 year Meunier (2014) [72]
`
`05 33%; one case of PTLD
`
`myelosuppressive
`mimic GVHD. Increased infections,
`diarrhea; can induce GI pathology changes that
`
`Usually well tolerated but can cause nausea and
`
`EBV reactivation/PTLD due to B cell depletion
`treatment. Increased risk for CMV; less risk for
`
`Effective GVHD prophylaxis; less studied for GVHD
`
`(2008) [71]
`
`Gomez-Almaguer
`
`ORR 83% (CR 33%)
`one case of PTLD
`higher TRM with ATG;
`33 vs. 24%, p = NS);
`
`one case of PTLD
`survival. 1 year OS 32%;
`and earlier. No impact on
`response for skin GVHD
`
`p = 0.11
`6 month OS 79 vs. 50%
`(CR 63 vs. 21%, p = 0.03);
`
`ORR 79 VS. 42%, p = 0.07
`
`ORR 54% (CR 20%). Best
`
`Day 28 response
`
`Day 30 response
`
`Day 28 response
`
`ORR 55 vs. 48%, p = NS (CR Van Lint (2006) [67]
`
`increased risk for PTLD
`to use; significant infectious complications,
`proof of efficacy; no consensus on which brand
`
`MacMillan (2002) [65] Commonly used agent for SR GVHD despite lack of
`
`Nishimoto (2015) [47]
`
`Safe procedure and no increase in infection rates.
`
`Burg/er (2014) [46]
`
`[40]
`
`Hackenbery (2007)
`
`radiology service is needed
`Pitfalls: trained and available interventional
`
`Response at discharge ORR 72%; 1.5 year OS 27% Weintraub (2010) [45]
`
`setting of GVHD and antacid therapy
`action of the drug, and not always optimal in
`suppression; pH milieu is important for effective
`absorption does occur with reports of adrenal
`
`Nonabsorbed is a misnomer, variable systemic
`
`Comments
`
`ORR 77% (CR 65.5%), 5 pts Castilla (2006) [44]
`
`Day 28 response
`
`Bertz (1999) [43]
`Author
`
`ORR 77 vs. 32%, p < 0.0,
`
`Results
`
`movement
`frequency of bowel
`
`p = 0.04
`1 year OS 71 vs. 58%,
`50 (31 vs. 48%, p = 0.12);
`BDP group through day
`Fewer treatment failure in
`
`ORR 83%; 1 year OS 50%
`
`systemic steroids
`50% pts required
`in CR relapsed; overall
`
`survival
`
`Day 28 response
`
`Retrospective
`
`GI)
`
`15 SR GVHD (12
`
`Day 28 response
`Day 28 response
`
`Retrospective
`Retrospective
`
`24 SR GI GVHD
`12 SR GI GVHD
`
`(purine analog)
`
`Pentostatin
`
`only, high dose MMF
`
`Retrospective, children Day 28 and 56
`
`with GI GVHD)
`14 SR GVHD (12
`
`Day 56 GVHD-free
`
`placebo
`MMF 1 g Q8 h VS.
`multicenter; PRD +
`placebo controlled,
`Phase Ill prospective,
`
`onset GVHD
`
`116 vs. 119 new
`
`ANC < 1500 excluded
`controls; pts NPO or
`with historical
`
`with GI)
`controls (11
`29 historical
`
`synthesis)
`(inhibitor of purine
`(MMF)
`
`19 (10 with GI) vs. Phase II prospective
`
`ANTI METABOLITES Mycophenolate mofetil
`
`Survival
`
`Retrospective
`
`24 SR GI GVHD
`
`Prospective, 2nd line
`
`with GI GVHD)
`
`antibody)
`emtuzumab (anti-CD52 18 SR GVHD, (8
`
`Al
`
`kg x 5 doses)
`days± ATG (1.25 mg/
`MP 5 mg/kg/d x 10
`211 pts; 61 with SR Phase 111, multicenter;
`
`(rabbit ATG)
`acute GVHD
`
`days+ GC
`15 mg/kg bid x 5
`1-5 courses ATG
`
`(horse ATG)
`lower GI.
`
`79 SR GVHD; 52% Retrospective; 2nd line; Day 28 response
`
`dose MP 180 mg
`19 SR GI GVHD vs. Phase 11, mean total
`
`controls
`14 historical
`
`Pilot study, ;,, 2nd line,
`
`)
`
`(MP 75 mg/m2
`
`11 SR GI GVHD
`
`Prospective, mean total Day 28 response
`
`12 SR GI GVHD
`
`dose MP 180 mg
`
`1 I diarrhea excluded
`placebo, pts with >
`1 mg/kg/d +
`PRD + BDP vs. PRD
`
`Phase 111, biopsy proven, Day 50 response;
`
`only
`with GI GVHD
`62 vs. 67 pts, all
`
`(ATG)
`
`Antithymocyte globulin
`
`(IAS)
`
`Intra-arterial steroids
`
`monotherapy BDP
`
`GI GVHD
`
`26 pts ,;; Grade 2 Prospective,
`
`proprionate (BOP)
`Oral beclomethasone
`
`di
`
`GC + BUD vs. GC only
`
`historical control
`GI GVHD vs. 19
`
`22 pts ;,, Grade 2 Retrospective, systemic Stool volume and
`End points
`Patient population
`
`Study design
`
`Budesonide (BUD)
`Agent
`
`AGENTS
`DEPLETING
`LYMPHOmE-
`
`Table 3. Nonspecific T-cell targeting.
`
`

`

`C:
`
`QJ .s
`"' %
`
`>(cid:173)
`"' Cl
`
`"' QJ
`cc
`
`QJ
`
`QJ
`
`> ·e
`5;(cid:173)
`g
`
`QJ
`Ct::
`
`C:
`C,
`·.;;
`QJ
`"O
`
`-C:
`
`QJ
`
`C, <
`
`EXPERT OPINION ON ORPHAN DRUGS 9 475
`
`horse- and rabbit-derived formulations of ATG in GVHD pro(cid:173)
`phylaxis is well established.[63,64] Rabbit-derived ATG induces
`more profound lymphopenia, but titers in the two brands vary
`widely and with no consensus as to which brand should be
`preferred. A retrospective series of 79 patients treated with
`horse ATG for SR GVHD showed a 54% CR/PR rate [65];
`patients in that study were more likely to have skin involve(cid:173)
`ment and less likely to have GI
`involvement compared to
`studies with less favorable results.[66] A large randomized
`study of rabbit ATG for SR GVHD did not improve outcomes,
`and its use was associated with a significantly increased risk of
`TRM.(67] Thus, improvements in response rates with ATG may
`not translate to survival benefit because of increased infec(cid:173)
`tious complications and Epstein-Barr virus (EBV)-associated
`lymphoproliferative disease.(68]
`Alemtuzumab, a humanized lgG monoclonal antibody to
`CD52 that is expressed on lymphocytes, monocytes, and APCs
`(Figure 1), is a more potent alternative to ATG,(69] and is
`effective as GVHD prophylaxis.(70] In small studies of SR
`GVHD patients, alemtuzumab provides overall response rates
`from 62 to 83%, with overall survival of 33-70%.(71,72] As
`expected, infectious complications frequently develop.
`
`6.6. Targeted therapies
`
`6.6.7. TNFa
`lnfliximab. lnfliximab (Figure 1) is a chimeric anti(cid:173)
`6.6. 7. 7.
`body that neutralizes TNFa and lyses the cells that produce it
`by binding to its membrane-bound form.(73] In retrospective
`series of patients with SR GVHD, infliximab produced overall
`response rates ranging from 15 to 60%, but these were asso(cid:173)
`ciated with high rates of fungal infections.(73-75] A rando(cid:173)
`mized, Phase Ill trial showed no benefit from the addition of
`infliximab to steroids as primary GVHD therapy.(76]
`
`6.6. 7 .2. Etanercept. Etanercept, which consists of
`two
`recombinant human TNFR (p75) monomers fused to the Fe
`portion of human lgG, neutralizes soluble TNFa (Figure 1), but
`does not lyse the cells producing it, and has a good safety
`profile.[77] In a prospective single-center study as primary
`GVHD therapy, the combination of etanercept and steroids
`was significantly superior to steroids alone (CR 69 vs. 33%);
`(30] however, these results were not reproduced in a rando(cid:173)
`mized, Phase II study.(56] The drug is not as effective in the SR
`GVHD setting, with response rates of 40-50%, including for
`patients with severe GI GVHD.[78,79]
`
`6.6.2. /L-6 receptor
`Tocilizumab is a humanized anti-lL6 receptor antibody that
`blocks IL-6 signaling, which is pivotal in the differentiation of
`CD4+/IL-17-secreting T (Th17) cells
`from na"i"ve T cells
`(Figure 1 ). In murine models, loss