Overexpressed differentiation antigens as targets of
`graft-versus-leukemia reactions
`Jeffrey J. Molldrem, MD, Krishna Komanduri, MD, and Eric Wieder, PhD
`
`The graft-versus-leukemia (GVL) effect associated with
`allogeneic blood and marrow transplantation has largely been
`a clinically described phenomenon until recently. We are
`beginning to understand the cellular and molecular nature of
`GVL, and in this review the authors highlight the potential for
`self-antigen–specific T lymphocytes to contribute to GVL. The
`authors focus on myeloid tissue–restricted proteins as GVL
`target antigens in CML and AML, and in particular on
`proteinase 3 and other azurophil granule proteins as targets
`for both autologous and allogeneic T-cell responses. Finally,
`the authors discuss myeloid self-antigen–directed alloreactivity
`in the context of our evolving understanding of the critical
`molecular determinants of allogeneic T-cell recognition. By
`altering T-cell receptor affinity, peptide specificity can be
`maintained and the potency of immunity can be enhanced in
`the MHC-mismatched setting. Curr Opin Hematol 2002, 9:503–508
`© 2002 Lippincott Williams & Wilkins, Inc.
`
`Transplantation Immunology Section, Department of Blood and Marrow
`Transplantation, University of Texas M. D. Anderson Cancer Center, Houston,
`Texas, USA.
`
`Correspondence to Jeffrey J. Molldrem, M.D., Chief, Transplantation Immunology
`Section, Department of Blood and Marrow Transplantation, University of Texas,
`M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Box 448 Houston,
`Texas 77030, USA; e-mail: jmolldre@notes.mdacc.tmc.edu
`
`Jeffrey J. Molldrem was supported by the U.S. Public Health Service (CA81247
`and CA85843) and by the Leukemia and Lymphoma Society of America
`(R6247–02).
`
`Current Opinion in Hematology 2002, 9:503–508
`
`Abbreviations
`
`CFU-GM
`CML
`CTL
`DLI
`GVHD
`GVL
`mHA
`MPO
`PR1-CTL
`Pr3
`TCR
`
`colony-forming unit granulocyte-macrophage
`chronic myelogenous leukemia
`cytotoxic T lymphocytes
`donor lymphocyte infusions
`graft-versus-host-disease
`graft-versus-leukemia
`minor histocompatibility antigen
`myeloperoxidase
`CTL specific for PR1
`proteinase 3
`T-cell receptor
`
`ISSN 1065–6251 © 2002 Lippincott Williams & Wilkins, Inc.
`
`The power of allogeneic lymphocytes to cure malignan-
`cies is perhaps best demonstrated by what happens to
`patients who receive donor lymphocyte infusions (DLI)
`for relapsed chronic myelogenous leukemia (CML). As
`many as 80% of these patients achieve a molecular re-
`mission [1,2], an effect that has been termed graft-
`versus-leukemia, or GVL. This is mediated mostly by
`T lymphocytes, because depletion of T cells from the
`graft abrogates this effect. Unfortunately, another T-cell–
`mediated effect, graft-versus-host-disease (GVHD), ac-
`companies DLI therapy in up to 50% of patients, thus
`limiting the full therapeutic potential of DLI. Because
`many patients achieve remission during flares of GVHD,
`it is uncertain whether GVL can be separated from
`GVHD or whether these phenomena are irrevocably
`linked. However, up to 55% of patients that do not de-
`velop GVHD also achieve molecular remission, suggest-
`ing that these immune reactions are separable [1,3]. If
`there were distinct effector cells or unique target anti-
`gens for the effector cells that produced GVL versus
`GVHD, then the full therapeutic potential of allogeneic
`DLI might be realized by treatment strategies that took
`advantage of these differences.
`
`Identifying differentiation antigens as
`GVL targets
`The range of target antigens for allogeneic donor lym-
`phocytes includes HLA molecules, minor histocompat-
`ibility antigens (mHAs), or self-antigens that are aber-
`rantly expressed in the tumor compared with normal
`tissues. In the case of HLA-matched BMT, alloreactivity
`directed against polymorphic mHA could account for
`both GVL and GVHD. Under these circumstances, the
`tissue distribution of the target mHA would direct the
`type of immune reaction. Certain mHA that have expres-
`sion restricted to the tissue from which the tumor is
`derived but not other host tissues might therefore also
`be ideal target antigens for preferential T-cell reactiv-
`ity (Fig. 1) against the tumor leading to graft-versus
`malignancy. This would require binding of the mHA to
`the HLA molecule with threshold recognition by T cells
`that have T-cell receptors (TCRs) that are specific for
`the recipient alternate polymorphism but not the donor
`polymorphism.
`
`Previous studies of immunity against solid tumors have
`revealed that most tumor antigens identified so far are
`nonmutated self-antigens that are aberrantly expressed
`in the tumor compared with normal host tissue [4••].
`DOI: 10.1097/01.MOH.0000032001.07903.35
`503
`Copyright© Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
`
`NOVARTIS EXHIBIT 2098
`Breckenridge v. Novartis, IPR 2017-01592
`Page 1 of 6
`
`

`

`504 Hematopoietic stem cell transplantation
`
`Figure 1. The spectrum of T-cell autoreactivity and alloreactivity
`
`The T-cell, pictured on top, engages peptide antigen in the
`context of MHC on the surface of an antigen-presenting
`cell. Different types of reactions may result, depending on
`whether the peptide or MHC are self-derived, allogeneic, or
`(regarding the peptide) foreign (that is, non-self and
`non-allogeneic). Published with permission [29].
`
`B,-M Peptide ••
`
`MHC-1 ...
`
`Self
`Non-self
`
`Immune
`
`A
`
`Response {
`Examples { alloMHC
`
`reactivity
`
`B
`X~Po,gg~iis or
`
`C
`
`Allogeneic
`
`Minor histocompatib ility
`antigens
`
`alloMHC
`reactivity
`
`D
`Auto logo us
`
`Autoimmunity
`
`Intracellular pathogens
`
`GVHD,GVL
`
`Cross-presented antigens
`
`HLA supertype
`response
`
`There are now several such examples in melanoma
`(MAGE, gp100, tyrosinase) and breast cancer (Her2/neu)
`[5]. An example of an aberrantly expressed tumor anti-
`gen in human leukemia is proteinase 3 (Pr3), a 26-kDa
`neutral serine protease that is stored in primary azurophil
`granules and is maximally expressed at the promyelocyte
`stage of myeloid differentiation [6–8]. Pr3 and two other
`azurophil granule proteins, neutrophil elastase and
`azurocidin, are coordinately regulated and the transcrip-
`tion factors PU.1 and C/EBP␣, which are responsible for
`normal myeloid differentiation from stem cells to mono-
`cytes or granulocytes, are important in mediating their
`expression [9]. In particular, PU.1 induces expression of
`the macrophage colony-stimulating factor receptor and
`the development of monocytes, whereas C/EBP␣ in-
`creases the expression of the granulocyte colony-
`stimulating factor receptor and leads to mature granulo-
`cytes [9,10]. These transcription factors have been
`implicated in leukemogenesis [10], and Pr3 itself may
`
`also be important in maintaining a leukemia phenotype
`because Pr3 antisense oligonucleotides halt cell division
`and induce maturation of the HL-60 promyelocytic leu-
`kemia cell line [11].
`
`We have also studied another myeloid-restricted protein,
`myeloperoxidase (MPO), a heme protein synthesized
`during very early myeloid differentiation that constitutes
`the major component of neutrophil azurophilic granules
`(Table 1). Produced as a single-chain precursor, my-
`eloperoxidase is subsequently cleaved into a light and
`heavy chain. The mature myeloperoxidase enzyme is
`composed of two light chains and two heavy chains [12]
`and produces hypohalous acids central to the microbici-
`dal activity of neutrophil. Importantly, MPO and Pr3 are
`both over-expressed in a variety of myeloid leukemia
`cells including 75% of CML patients, approximately
`50% of acute myeloid leukemia patients, and approxi-
`mately 30% of myelodysplastic syndrome patients [13].
`
`Table 1. Myeloid proteins as potential tissue-restricted leukemia antigens
`
`Protein
`
`Chromosome
`
`Proteinase 3*
`neutrophil Elastase
`Myeloperoxidase
`Cathepsin G*
`
`19p
`19p
`17q22
`14q11.2
`
`mRNA
`
`Normal
`CD34+
`
`Leukemic
`CD34+
`
`Autoimmune
`syndrome
`
`−/+
`−
`+
`−
`
`+
`+
`++
`+
`
`Wegener’s
`Wegener’s & Vasculitis
`Vasculitis
`Sclerosing cholangitis
`
`*Naturally processed and presented by CML blasts.
`CML, chronic myeloid leukemia.
`Data from Barrett et al. [29].
`
`Copyright© Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
`
`NOVARTIS EXHIBIT 2098
`Breckenridge v. Novartis, IPR 2017-01592
`Page 2 of 6
`
`

`

`Differentiation antigens as targets of GVL reactions Molldrem et al. 505
`
`What may be critical for our ability to identify T-cell
`antigens in these proteins is the observation that Pr3 is
`the target of autoimmune attack in Wegener’s granulo-
`matosis [14] and MPO is the target antigen in small
`vessel vasculitis [12,15,16]. There is evidence for both
`T-cell and humoral immunity in patients with these dis-
`eases. Wegener’s granulomatosis is associated with pro-
`duction of cytoplasmic antineutrophil cytoplasmic anti-
`bodies with specificity for Pr3 [17], whereas microscopic
`polyangiitis and Churg-Strauss syndrome are associated
`with the production of perinuclear ANCA antibodies
`with specificity for MPO [18,19]. T cells taken from af-
`fected individuals proliferate in response to crude ex-
`tracts from neutrophil granules and to the purified pro-
`teins [15,20]. These findings suggest that T-cell
`responses against these proteins might be relatively easy
`to elicit in vitro using a deductive strategy to identify
`HLA-restricted peptide epitopes. Based on this hypoth-
`esis, we identified PR1, an HLA-A2.1–restricted non-
`amer derived from Pr3, as a leukemia-associated antigen
`[21•,22–24] by first searching the length of the protein
`using the HLA-A2.1 binding motif, the most common
`HLA allele. Peptides predicted to have high-affinity
`binding to HLA-A2.1 were synthesized, confirmed to bind,
`and then used to elicit peptide-specific cytotoxic T lym-
`phocytes (CTL) in vitro from healthy donor lymphocytes.
`
`We have found that PR1 can be used to elicit CTL from
`HLA-A2.1+ normal donors in vitro, and that T-cell im-
`munity to PR1 is present in healthy donors and in many
`patients with CML that are in remission. These PR1-
`specific CTL show preferential cytotoxicity toward
`allogeneic HLA-A2.1+ myeloid leukemia cells over
`HLA-identical normal donor marrow [22]. In addition,
`PR1-specific CTLs inhibit colony-forming unit granulo-
`cyte-macrophage (CFU-GM) from the marrow of CML
`patients, but not CFU-GM from normal HLA-matched
`donors [23], suggesting that leukemia progenitors are
`also targeted.
`
`Using PR1/HLA-A2 tetramers to detect CTL specific for
`PR1 (PR1-CTL), we found a significant correlation with
`cytogenetic remission after treatment with interferon-␣
`and the presence of PR1-CTL [21•]. Somewhat surpris-
`ingly, PR1-CTLs were also identified in the peripheral
`blood of some allogeneic transplant recipients who
`achieved molecular remission and who had converted to
`100% donor chimerism. PR1/HLA-A2 tetramer-sorted
`allogeneic CTL from patients in remission were able to
`kill CML cells but not normal bone marrow cells in
`4-hour cytotoxicity assays, thus demonstrating that the
`PR1 self-antigen is also recognized by allogeneic CTL
`[21•]. These studies have established PR1 as a human
`leukemia-associated antigen, and they established that
`PR1-specific CTLs contribute to the elimination of
`CML [21•].
`
`Recently we found another peptide, MY4, a 9-amino-
`acid peptide derived from MPO that binds to HLA-A2.1,
`which can be used to elicit CTL from HLA-A2.1+
`normal donors in vitro [25]. MY4-specific CTLs show
`preferential cytotoxicity toward allogeneic HLA-A2.1+
`myeloid leukemia cells over HLA-identical normal do-
`nor marrow [25]. MY4-specific CTLs also inhibit colony-
`forming unit granulocyte-macrophage (CFU-GM)
`from the marrow of CML patients but not CFU-GM
`from normal HLA-matched donors. Like PR1, MY4 is
`therefore a peptide antigen that can elicit leukemia-
`specific CTL.
`
`Because of the many striking similarities between im-
`munity to Pr3 and to MPO, it is likely that similar meth-
`ods applied to the study of immunity against MPO-
`derived peptides will establish MY4 and other peptides
`as important leukemia-associated antigens [26]. Using a
`deductive strategy to uncover potential tumor antigens,
`we are currently studying sequence data from the human
`genome project to determine other HLA-restricted epi-
`topes from tissue-restricted antigens. There is a high
`likelihood that other peptide epitopes can be deter-
`mined using this approach, especially by focusing on
`those proteins that are already the known targets of
`T-cell–mediated autoimmunity.
`
`T-cell receptor affinity influences GVL
`More recently, we have shown that distinct populations
`of PR1-CTL with either high or low TCR affinity for
`PR1 can be elicited from PBMC of healthy donors. The
`high-affinity PR1-CTL cause higher specific lysis of
`CML cells than low-affinity PR1-CTL. Interestingly, we
`also found that when high-affinity PR1-CTLs were ex-
`posed to target cells that expressed high concentrations
`of target antigens, the PR1-CTL underwent apoptosis
`within 18 hours. However, there was no apoptosis when
`the high-affinity PR1-CTLs were exposed to a 2-log
`lower concentration of PR1 antigen. Furthermore, we
`have been unable to either detect or elicit high-affinity
`PR1-CTL in vitro from PBMC of untreated CML
`patients. Because healthy HLA-A2+ individuals have
`PR1-CTL with high-affinity TCR, however, this sug-
`gests that the high-affinity PR1-CTL may have been
`deleted during the outgrowth of the leukemia by CML
`cells that over-express the PR1 tumor antigen.
`
`Taken together these findings suggest that, in addition
`to HLA disparities and polymorphic mHAs, self-antigens
`may be the targets of alloreactive T cells. These obser-
`vations form the basis for a mechanism of alloreactivity
`and subsequent new treatment strategies based on tar-
`geting self-antigens in the allogeneic setting. Specifi-
`cally, GVL alloreactivity may in part be caused by the
`transfer from donor to recipient of high-affinity CTL
`with leukemia self-antigen specificity that were not de-
`leted from the T-cell repertoire during normal T-cell
`
`Copyright© Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
`
`NOVARTIS EXHIBIT 2098
`Breckenridge v. Novartis, IPR 2017-01592
`Page 3 of 6
`
`

`

`506 Hematopoietic stem cell transplantation
`
`development in the donor. On this basis, GVL could be
`separated from GVHD if the target self-antigen expres-
`sion was limited to hematopoietic tissue only. Further
`specificity from aberrant expression of the target self-
`antigen in the leukemia compared with normal hemato-
`poietic cells might give rise to a critical number of rec-
`ognizable surface peptide epitopes that would surpass
`the activation threshold of high-affinity T cells, whereas
`the lower level of antigen expressed in the normal he-
`matopoietic cells would not. This would result in pref-
`erential killing and elimination of leukemia cells over
`normal hematopoietic cells by the transplanted high-
`affinity donor T cells. As a consequence, residual normal
`recipient hematopoietic cells would be spared and could
`then coexist with donor hematopoietic cells after suc-
`cessful elimination of the leukemia, a phenomenon that
`occurs in some BMT recipients that achieve cytogenetic
`remission.
`
`Arguing against this hypothesis is the observation that
`CML recipients of syngeneic stem cell grafts, which
`have few mHA differences but which should also contain
`high-affinity PR1-CTL, suffer higher relapse rates than
`do recipients of allogeneic grafts [27]. However, because
`high-affinity PR1-CTLs are present at a very low pre-
`cursor frequency in healthy donors, major and minor his-
`tocompatibility antigenic differences may be required to
`provide generalized heightened immunity via indirect
`effects mediated by cytokine secretion, which might
`broadly decrease the threshold of TCR activation and
`drive the expansion of high-affinity self-antigen–specific
`T cells. This would also explain the development of
`GVHD, because this could lead to the uncovering of
`cryptic antigens and also to epitope spreading [28••].
`More effective GVL might therefore be observed after
`syngeneic BMT if higher numbers of high-affinity CTL
`were initially transplanted. Consistent with this is the
`clinical observation that fewer relapses occur in synge-
`neic graft recipients who receive higher total nucleated
`cell doses during initial transplant [29], suggesting that
`an initially high number of high-affinity self-antigen–
`specific CTL might compensate for their innately low
`precursor frequency and the absence of significant allo-
`reactivity in this setting.
`
`Molecular basis of allogeneic GVL
`against self-antigens
`The observation that self-antigens can also be recog-
`nized as tumor antigens by allogeneic T cells presents an
`opportunity to redirect potent alloreactivity toward these
`self-antigens. Our observations, which are consistent
`with an evolving overall understanding of the molecular
`basis of allorecognition, suggest a unique approach to
`immunotherapy. It has long been recognized that very
`vigorous T-cell responses occur when donor tissue is
`transplanted into an MHC-mismatched recipient, where
`up to 10% of recipient peripheral T cells respond to
`
`allo-MHC antigen. This high frequency of recipient-
`reactive donor T cells occurs because of the increased
`binding energy of donor TCR to the recipient
`peptide/allo-MHC combination, and either the peptide
`or the polymorphic amino acid differences in the allo-
`MHC may account for this higher binding energy. In
`addition, either interaction may increase the binding
`energy relative to that of donor TCR bound to peptide
`plus donor (self)-MHC. Although it was originally
`thought that allo-MHC differences accounted for the in-
`creased binding energy, Reiser et al., recently showed
`that T-cell alloreactivity can be caused by more effective
`interaction of the TCR with both peptide and allo-MHC
`residues [30••].
`
`TCR on the surface of CD8+ CTL recognize short pep-
`tides 8 to 11 amino acids long that are derived from
`intracellular proteins and bind to MHC class I. During
`normal T-cell maturation, TCRs are selected based on
`their binding affinity to peptide plus self-MHC, a pro-
`cess referred to as positive selection [31]. Likewise, an-
`tigen recognition by alloreactive T cells also depends on
`peptides within the allo-MHC groove. Most of these T
`cells exhibit some degree of peptide specificity, and the
`frequency of peptide-specific alloreactive T cells was re-
`cently found to be higher when the allo-MHC was more
`similar to self-MHC [30]. Thus, polymorphic residues on
`allo-MHC might give rise to altered amino acids that
`could raise the binding threshold of the TCR above the
`interactions produced by shared residues on self-MHC,
`the latter having been accomplished through positive T-
`cell selection in the donor. Consequently, an allo-MHC
`molecule with more extensive polymorphism would
`have a higher likelihood of losing the energy of interac-
`tions gained from positive selection, and T cells that can
`react productively with these highly polymorphic allo-
`MHC would be of lower frequency than T cells that
`have the potential to cross-react with allo-MHC of a
`lower degree of polymorphism.
`
`For the T cell to become activated, the added TCR
`interaction with the bound peptide need only raise the
`affinity slightly above the energy contributed by the
`TCR interaction with allo-MHC alone. The observation
`that only small increases in binding energy above the
`direct contribution by TCR interaction with allo-MHC
`are necessary to reach threshold for T-cell activation is
`consistent with the observation that alloreactive T cells,
`although peptide-dependent, appear to be less peptide-
`specific than TCR interactions with self. This decreased
`peptide specificity refers only to T-cell activation, a
`downstream measure of antigen recognition and cell
`function. The crystallographic data from Reiser et al. is
`consistent with the likelihood that various peptides,
`when bound to an allo-MHC, may appear cross-reactive
`in eliciting T-cell responses because their interactions
`with the TCR are above the critical threshold of activa-
`
`Copyright© Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
`
`NOVARTIS EXHIBIT 2098
`Breckenridge v. Novartis, IPR 2017-01592
`Page 4 of 6
`
`

`

`Differentiation antigens as targets of GVL reactions Molldrem et al. 507
`
`tion but their individual affinities for the TCR may be
`different [30]. This has recently been demonstrated in a
`murine model [32] and in humans, where non-self CTLs
`maintain specificity for the HA-1 minor histocompatibil-
`ity antigen across different HLA alleles [33•].
`
`The MHC alleles can differ from one another by as many
`as 20 amino acids, and most of these polymorphic resi-
`dues line the peptide binding cleft that determine pep-
`tide-binding specificity. Nevertheless, a few polymor-
`phic residues are exposed on the outer surface of the
`MHC ␣-helices and hence would be able to interact with
`the TCR. These might allow alloreactive TCR to adopt
`an MHC-binding geometry that is similar to the original
`TCR conformation that contacted self-MHC molecules
`and resulted in positive T-cell selection. In addition,
`changes in peptide/MHC shape complementarity might
`also occur by buried or non-exposed polymorphisms in
`the MHC that preserve peptide specificity but that may
`still increase TCR-binding affinity [30].
`
`To better understand how to maximize the full thera-
`peutic potential of alloreactive T cells, we must consider
`the degeneracy of a single peptide binding to various
`MHC alleles. Distinct MHC alleles that bind a single
`common peptide have been termed super-type alleles,
`and they share similar amino acid residues in their pep-
`tide-binding pockets that bind common peptides [34,35].
`However, polymorphic residues on the ␣1 and ␣2 do-
`mains of super-type allo-MHC may also contribute
`higher binding energies with alloreactive TCR than resi-
`dues at the same positions on self-MHC. Under this
`circumstance, the bound peptide would become the
`common TCR-restricting element, directing the potent
`effector function of the alloreactive CTL against target
`cells that express the same common peptide but distinct
`polymorphic allo-MHC. If the peptide were preferen-
`tially expressed in the tissue from which the tumor was
`derived, it would be transformed into a potent tumor
`antigen in the context of this alloreactivity.
`
`This model suggests how alloreactivity directed toward
`tissue-restricted self-peptides might be exploited to
`take advantage of the vigorous alloreactivity that occurs
`after MHC-mismatched stem cell transplantation. In the
`case of the PR1 peptide, for example, we have recently
`shown that lymphocytes from an HLA-A*0201-positive
`healthy donor contain a population of T cells with high-
`affinity PR1-specific TCR. We have also found that PR1
`also binds equally well to other alleles in the HLA-A2
`super type (Molldrem, unpublished observations, June
`2002). Therefore, CTL adoptively transferred from an
`HLA-A*0201 donor to a HLA-A*0205 CML patient
`might result in more potent GVL against CML if the
`TCR of the donor CTL reached activation threshold
`earlier than residual autologous PR1-CTL similarly ex-
`posed to PR1 in the context of HLA-A*0205. Because
`
`CTL with high-affinity TCR for the PR1 peptide exist
`in most healthy donors without evidence of self-
`hematopoietic tissue destruction, it is reasonable to be-
`lieve that normal HLA-A*0205 hematopoietic cells ex-
`pressing normal
`levels of PR1 also would not be
`recognized by the alloreactive HLA-A*0201 CTL with
`higher TCR affinity for HLA-A*0205. This would facili-
`tate the development of hematopoietic microchimerism
`in the transplant recipient with both HLA-A*0205 and
`HLA-A*0201 hematopoietic cells.
`
`Advantages and disadvantages of
`self-antigens as GVL targets
`Several authors have suggested that one way to enhance
`GVL and reduce GVHD would be to adoptively transfer
`antigen-specific T cells from the donor to the recipient
`[36••,37]. Adoptive transfer to BMT recipients of allo-
`reactive T cells with specificity for self-peptides after an
`initial T-cell–depleted MHC-mismatched transplant of-
`fers several potential advantages over strategies utilizing
`precise HLA and possible mHA matching to reduce the
`incidence of GVHD. First, it would greatly expand the
`number of potential donors for allogeneic stem cell trans-
`plantation, which is the largest obstacle to extending this
`potentially curative treatment modality to more patients.
`Donor-recipient pairs that shared a common HLA super-
`type would be sufficient. Second, the time required to
`expand peptide antigen-specific CTL ex vivo for adop-
`tive transfer to recipients to induce GVL might be elimi-
`nated or greatly reduced because of the high initial pre-
`cursor frequency of the alloreactive CTL. Third,
`because the target peptide is a self-antigen, it would
`eliminate the need to find tissue-restricted mHA differ-
`ences between donor and recipient if mHA-specific
`CTL were to be adoptively transferred to the recipient
`to induce GVL reactivity.
`
`This strategy of self-peptide–directed alloreactivity
`might also be applied to the treatment of solid tumors,
`where many self-antigens have already been discovered
`but where effective autologous immune responses are
`lacking [38]. It also suggests a possible future strategy for
`the treatment of autoimmune diseases if suitable peptide
`antigens could be identified and their gene expression
`was restricted to T cells or even to hematopoietic cells.
`
`Conclusions
`Obstacles to this approach remain, however. We must
`
`(2)
`
`(1) determine the key MHC residues that are involved
`in positive selection;
`identify certain tissue-restricted self-peptides that
`are recognized by T cells; and
`(3) determine which of those peptides also bind to dif-
`ferent alleles that are confined to a given HLA
`super type.
`
`Copyright© Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
`
`NOVARTIS EXHIBIT 2098
`Breckenridge v. Novartis, IPR 2017-01592
`Page 5 of 6
`
`

`

`20 Ballieux BE, van der Burg SH, Hagen EC, et al.: Cell-mediated autoimmunity
`in patients with Wegener’s granulomatosis (WG). Clin Exp Immunol 1995,
`100:186–193.
`1 Molldrem JJ, Lee PP, Wang C, et al.: Evidence that specific T lymphocytes
`may participate in the elimination of chronic myelogenous leukemia. Nat Med
`2000, 6:1018–1023.
`Establishes proteinase 3 as a human leukemia-associated antigen and as a target
`for allogeneic T lymphocytes.
`22 Molldrem J, Dermime S, Parker K, et al.: Targeted T-cell therapy for human
`leukemia: cytotoxic T lymphocytes specific for a peptide derived from protein-
`ase 3 preferentially lyse human myeloid leukemia cells. Blood 1996,
`88:2450–2457.
`23 Molldrem JJ, Clave E, Jiang YZ, et al.: Cytotoxic T lymphocytes specific for a
`nonpolymorphic proteinase 3 peptide preferentially inhibit chronic myeloid
`leukemia colony-forming units. Blood 1997, 90:2529–2534.
`24 Molldrem JJ, Lee PP, Wang C, et al.: A PR1-human leukocyte antigen-A2
`tetramer can be used to isolate low- frequency cytotoxic T lymphocytes from
`healthy donors that selectively lyse chronic myelogenous leukemia. Cancer
`Res 1999, 59:2675–2681.
`25 Braunschweig I, Wang C, Molldrem J: Cytotoxic T lymphocytes (CTL) spe-
`cific for myeloperoxidase-derived HLA-A2-restricted peptides specifically
`lyse AML and CML cells. Blood 2000, 96:3291.
`26 Kochenderfer JN, Molldrem JJ: Leukemia Vaccines. Curr Oncol Rep 2001,
`3:193–200.
`27 Gale RP, Horowitz MM, Ash RC, et al.: Identical-twin bone marrow trans-
`plants for leukemia. Ann Intern Med 1994, 120:646–652.
`28 Anderton SM, Wraith DC: Selection and fine-tuning of the autoimmune T-cell
`repertoire. Nat Rev Immunol 2002, 2:487–498.
`(cid:127)(cid:127)
`The mechanisms of peripheral T-cell tolerance are discussed and the current mod-
`els of the development of autoimmunity are reviewed.
`29 Barrett AJ, Ringden O, Zhang MJ, et al.: Effect of nucleated marrow cell dose
`on relapse and survival in identical twin bone marrow transplants for leukemia.
`Blood 2000, 95:3323–3327.
`30 Reiser R, Darnault C, Guimezanes A, et al.: Crystal structure of a T-cell re-
`ceptor bound to an allogeneic MHC molecule. Nat Immunology 2000,
`(cid:127)(cid:127)
`1:291–297.
`The crystal structure data of an allogeneic immune synapse is presented that
`shows both the peptide and the MHC are critical for determining T-cell alloreactiv-
`ity.
`31 Alam SM, Travers PJ, Wung JL, et al.: T-cell-receptor affinity and thymocyte
`positive selection. Nature 1996, 381:616–620.
`
`(cid:127)2
`
`32
`
`Luz JG, Huang M, Garcia KC, et al.: Structural comparison of allogeneic and
`syngeneic T-cell receptor-peptide-major histocompatibility complex com-
`plexes: a buried alloreactive mutation subtly alters peptide presentation sub-
`stantially increasing V(beta) Interactions. J Exp Med 2002, 195:1175–1186.
`3 Mutis T, Blokland E, Kester M, et al.: Generation of minor histocompatibility
`antigen HA-1-specific cytotoxic T cells restricted by nonself HLA molecules:
`a potential strategy to treat relapsed leukemia after HLA-mismatched stem
`cell transplantation. Blood 2002, 100:547–552.
`This paper demonstrates an example of preserved peptide specificity across HLA
`barriers for alloreactive T cells.
`34 Bertoni R, Sidney J, Fowler P, et al.: Human histocompatibility leukocyte an-
`tigen-binding supermotifs predict broadly cross-reactive cytotoxic T lympho-
`cyte responses in patients with acute hepatitis. J Clin Invest 1997, 100:503–
`513.
`
`(cid:127)3
`
`35
`
`del Guercio MF, Sidney J, Hermanson G, et al.: Binding of a peptide antigen
`to multiple HLA alleles allows definition of an A2-like supertype. J Immunol
`1995, 154:685–693.
`36 Appelbaum FR: Haematopoietic cell transplantation as immunotherapy.
`Nature 2001, 411:385–389.
`(cid:127)(cid:127)
`This review highlights the use of BMT as a platform for cellular-based immuno-
`therapy and for post-BMT vaccination strategies to treat hematological malignan-
`cies.
`37 Barrett J, Jiang Y-Z: Allogeneic immunotherapy for malignant diseases. New
`York: Marcel Dekker; 2000.
`
`38
`
`Lee PP, Yee C, Savage PA, et al.: Characterization of circulating T cells spe-
`cific for tumor-associated antigens in melanoma patients. Nat Med 1999,
`5:677–685.
`
`508 Hematopoietic stem cell transplantation
`
`In the future, allogeneic stem cell transplantation is
`likely to evolve as a platform for delivering antigen-
`specific adoptive cellular therapy involving the trans-
`fer of alloreactive T cells with the appropriate antigen
`specificity.
`
`References and recommended reading
`Papers of particular interest, published within the annual period of review,
`have been highlighted as:
`(cid:127)
`Of special interest
`(cid:127)(cid:127)
`Of outstanding interest
`1
`
`Giralt SA, Kolb HJ: Donor lymphocyte infusions. Curr Opin Oncol 1996,
`8:96–102.
`
`2
`
`3
`
`Kolb HJ, Schattenberg A, Goldman JM, et al.: Graft-versus-leukemia effect of
`donor lymphocyte transfusions in marrow grafted patients. European Group
`for Blood and Marrow Transplantation Working Party Chronic Leukemia.
`Blood 1995, 86:2041–2050.
`
`Kolb HJ, Holler E: Adoptive immunotherapy with donor lymphocyte transfu-
`sions. Curr Opin Oncol 1997, 9:139–145.
`
`4
`Pardoll DM: Spinning molecular immunology into successful immunotherapy.
`Nature Rev Immunol 2002, 2:227–238.
`(cid:127)(cid:127)
`The difficulties of immunotherapy and the current laboratory and clinical ap-
`proaches to address those difficulties are discussed. This is an excellent brief over-
`view of the field of tumor immunology.
`5
`
`Boon T, Coulie PG, Van den Eynde B: Tumor antigens recognized by T cells.
`Immunol Today 1997, 18:267–268.
`
`6
`
`7
`
`8
`
`Sturrock AB, Franklin KF, Rao G, et al.: Structure, chromosomal assignment,
`and expression of the gene for proteinase 3. J Biol Chem 1992, 267:21193.
`
`Chen T, Meier R, Ziemiecki A, et al.: Myeloblastin/proteinase 3 belongs to the
`set of negatively regulated primary response genes expressed during in vitro
`myeloid differentiation. Biochem Biophys Res Commun 1994, 200:1130–
`1135.
`
`Muller-Berat N, Minowada J, Tsuji-Takayama K, et al.: The phylogeny of pro-
`teinase 3/myeloblastin, the autoantigen in Wegener’s granulomatosis, and
`myeloperoxidase as shown by immunohistochemical studies on human leu-
`kemic cell lines. Clin Immunol Immunopathol 1994, 70:51–59.
`
`9
`
`Zhang P, Nelson E, Radomska HS, et al.: Induction of granulocytic differen-
`tiation by 2 pathways. Blood 2002, 99:4406–4412.
`10 Behre G, Zhang P, Zhang DE, et al.: Analysis of the modulation of transcrip-
`tional activity in myelopoiesis and leukemogenesis. Methods 1999, 17:231–
`237.
`11 Bories D, Raynal MC, Solomon DH, et al.: Down-regulation of a serine pro-
`tease, myeloblastin, causes growth arrest and differentiation of promyelocytic
`leukemia cells. Cell 1989, 59:959.
`12 Borregaard N, Cowland JB: Granules of the human neutroph