1ST CUNEO CITY IMMUNOTHERAPY CONFERENCE 17-18 Maggio 2018 PASSIVE IMMUNOTHERAPY ANTIBODY-DRUG CONJUGATES GEMTUZUMAB OZOGAMICIN

Do. Filippo Gherlinzoni Direore Unità Operava Complessa di Ematologia Ospedale “Ca’ Foncello” - Treviso IL RIECCOLO Blood Reviews 28 (2014) 143–153

Contents lists available at ScienceDirect

Blood Reviews

journal homepage: www.elsevier.com/locate/blre

REVIEW The past and future of CD33 as therapeutic target in acute myeloid leukemia

a a,b a,b,c, George144 S. Laszlo , Elihu H. Estey , Roland B. Walter ⁎G.S. Laszlo et al. / Blood Reviews 28 (2014) 143–153 a Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA b Department of Medicine, Division of Hematology, University of Washington, Seattle, WA, USA c Department of Epidemiology, University of Washington, Seattle, WA, USA as cytokine-induced cellular proliferation [14,33]. Engagement with CD33 antibodies has similarly been shown to affect cytokine and article info abstract chemokine secretion by monocytes, although both increases [34] or

Keywords: CD33 is a myeloid differentiation antigen with endocytic properties.decreases It is broadly[33] expressedhave on been acute myeloidobserved in vitro. Acute myeloid leukemia leukemia (AML) blasts and, possibly, some leukemic stem cells and has therefore been exploited as target for Antibody therapeutic antibodies for many years. The improved survival seen in many patients when the antibody-drug Antibody-drug conjugate conjugate, gemtuzumab ozogamicin, is added to conventional chemotherapy validates this approach. However, fi 2.3. Endocytic properties of CD33 Bispeci c antibody many attempts with unconjugated or conjugated antibodies have been unsuccessful, highlighting the challenges BiTE of targeting CD33 in AML. With the development of improved immunoconjugates and CD33-directed strategies CD33 fi Chimeric antigen receptor that harness immune effector cells, therapeutics with enhanced ef cacy may soon become available. Toxic effects A therapeuticallyfi important characteristic of CD33 is its internaliza- Gemtuzumab ozogamicin on normal hematopoietic cells may increase in parallel with this increased ef cacy and demand new supportive Immunotherapy care measures, including possibly rescue with donor cells, to minimizetion when morbidity bound and mortality by bivalent from drug- antibodies; this process is slower than that Radioimmunoconjugate induced cytopenias and to optimize treatment outcomes with these agents in patients with AML. observed© 2014 Elsevier with other Ltd. All rights cell surface reserved. antigens such as the transferrin recep- tor [35–43]. Mechanistic studies indicate that endocytosis is largely lim- ited and determined by the intracellular domain of CD33 and may be regulated by tyrosine phosphorylation and perhaps ubiquitylation of 1. Introduction 2. Physiologic characteristics of CD33 the cytoplasmic tail, although some internalization of CD33/antibody Since the invention of hybridoma technology 4 decades ago, mono- CD33 is a member of the sialiccomplexes acid-binding occursimmuno in aglobulin-like phosphorylation-independent manner [41–43]. clonal antibodies have revolutionized the care of patients with cancer. lectins (Siglecs), a discrete subset of the immunoglobulin (Ig) super- An increasing number of unconjugated, toxin-loaded, and radiolabeled family molecules (Fig. 1) [4,5]Related. This 67kD to single this pass endocytic transmembrane property, engagement of CD33 with bivalent antibodies have shown anti-tumor efficacy and have been approved glycoprotein is characterized byantibodies an amino-terminal leads V-set to a Ig-like decrease domain (“modulation”) of CD33 cell surface levels for indications in an expanding list of malignancies, including acute that mediates sialic acid binding and a C2-set Ig-like domain in its extra- myeloid leukemia (AML) [1,2]. AML has been a paradigm for the thera- cellular portion [6–8]. Alternative[35,40,44] splicing. Although of CD33 RNA new leads CD33 to a sites are continuously expressed [40], peutic use of monoclonal antibodies, in no small part because malignant shorter isoform that is expressedthis on feature the cell surface. could This reduce isoform the lacks efficacy of CD33-directed therapeutics cells are readily accessible and express well-defined cell surface anti- the V-set Ig-like domain as well as the disulfide bond linking the gens. Most efforts to date have focused on exploiting CD33 as a target V- and C2-set Ig-like domains.because While the of biological the reduction relevance of of available this target binding sites. in this disease, and the CD33-directed immunoconjugate, gemtuzumab splicing process is unknown, it may be important for the development ozogamicin (GO), was the first anti-cancer antibody-drug conjugate to and use of CD33-directed drugs. Specifically, a dominant epitope, recog- obtain marketing approval in the U.S. [3] Still, targeting CD33 has nized by the majority of initial3. CD33 CD33 antibodies, in AML is located and other on the malignanciesV-set proven challenging, as perhaps best reflected by the eventual market Ig-like domain. Thus, some CD33 antibody-based therapeutics will only withdrawal of GO because of concerns over excess toxicity and lack of recognize the full-length but not the shorter splice isoform of CD33 fi efficacy. In this article, we will summarize the biologic characteristics [9,10]. Depending on how antigen positivity is de ned, CD33 is found on at of CD33, emphasizing the properties that make it appealing as a thera- The cytoplasmic tail of CD33least contains a subset 2 conserved of blasts tyrosine-based in nearly all AML patients [45,46], consistent with peutic target, appraise attempts made thus far with CD33-directed ther- inhibitory signaling motifs, which, upon phosphorylation by Src family its characteristic as a myeloid differentiation antigen. Although surface apies, andFig. discuss 1. Structure current and of CD33. future Scheme therapeutic depicting directions the domain in this fi structureeld. kinases, of CD33 provide as well docking as indi- sites for the recruitment and activation of Src Laszlo G.S. et al Blood Reviews 2014, 143-153 homology-2 (SH2) domain-containing tyrosine phosphatases such as vidual amino acids that have been implicated in phosphorylation or ubiquitylation events. levels show considerable inter-patient variability (N2-log fold) [15,45, SHP-1 and SHP-2. While the signaling events downstream of CD33 4 Abbreviations: C2, C2-set Ig-like domain; P, phospho-; SFKs, Src-familyremain kinases; poorly understood,Ub, ubiq- these46] phosphatases, CD33 expression may not only is dephos- relatively limited with an average of ~10 ⁎ Corresponding author at: Clinical Research Division, Fred Hutchinson Cancer Research uitin; V, V-set Ig-like domain. phorylate CD33 as part of a negative feedback loop but also dephos- Center; 1100 Fairview Ave N, D2-190; Seattle, WA 98109-1024, USA. Tel.: +1 206 667 molecules/AML blast [15,47] and is typically even lower in immature 3599; fax: +1 206 667 5255. phorylate and negatively regulate nearby receptors+ [11- –13]. Through+ + - E-mail address: [email protected] (R.B. Walter). its SH2 domain, suppressor(e.g. of cytokine CD34 signaling/CD38 3/CD123 (SOCS3) canor CD34 /CD38 ) cell subsets [46,48]. compete with SHP-1 or SHP-2 for binding to phosphorylated CD33 and CD33 expression correlates with disease characteristics: for example, http://dx.doi.org/10.1016/j.blre.2014.04.001recruit the ECS (Elongin B/C-Cul2/Cul5-SOCS-box protein) E3 ubiquitin expression is homogeneous and typically bright in acute promyelocytic 0268-960X/©ligase 2014 Elsevier complex, Ltd. All rights which reserved. ultimately leads to ubiquitylation and results in leukemias (APL) [49,50]. High levels of CD33 are also associated with accelerated proteasomal degradation of CD33 (Fig. 1) [14]. NPM1 as well as high allelic FLT3/ITD mutations, while expression is generally low with core-binding factor translocations [45,46,51]. Likely 2.1. CD33 as myeloid differentiation antigen at least partially related to such associations, it has been a recurrent observation in pediatric trials that, on average, patients whose AML In healthy individuals, CD33 is primarily a myeloid differentiation blasts display high CD33 levels experience inferior disease-free and antigen with initial expression at the very early stages of myeloid cell overall survival when treated with conventional chemotherapy that development: it is found on normal multipotent myeloid precursors, does not include CD33-targeted agents [52,53]; similar data from unipotent colony-forming cells, and maturing granulocytes and mono- adult patients is currently not available. In AML patients, CD33 can cytes. On CD34+/CD33+ bone marrow cells, CD33 expression has also be detected as soluble protein in the circulation and may provide been estimated to average around 8 × 103 molecules/cell, although some prognostic information [54,55]; however, its role as predictive levels vary widely (1–20 × 103 molecules/cell) [15]. Expression is biomarker has not been studied in detail. It is also unclear as to what down-regulated to lower levels on neutrophils (~2–2.5 × 103 mole- degree, if any, soluble CD33 might interfere with the therapeutic effica- cules/cell) and macrophages but retained on circulating monocytes cy of CD33 antibodies, although some in vitro evidence suggests that and dendritic cells [15–20]. CD33 can also be displayed on subsets of soluble CD33 may not impact the activity of CD33-targeted immuno- B-cells and activated T- and natural killer (NK) cells [9,21–27]. In con- therapy [56]. trast, CD33 is not expressed outside the hematopoietic system or on Besides its broad expression on AML blasts, a main impetus to pluripotent hematopoietic stem cells, with the latter indicated both by pursue CD33 as a drug target came from the notion that the stem cells in vitro studies of normal bone marrow [18,19,28] and by the delayed underlying some AMLs could be CD33+, implying that CD33-directed but durable multi-lineage engraftment after transplantation of CD33- therapy could potentially eradicate malignant stem and/or progenitor depleted autografts in patients with AML [29,30]. cells in such cases while sparing normal hematopoietic stem cells [57, 58]. This possibility was first indicated by classic X chromosome inacti- 2.2. Putative functions of CD33 vation studies, which found clonal dominance limited to granulocytes and monocytes in a subset of leukemias, suggesting that expansion of Increasing evidence suggests a role for CD33 and related Siglecs in the malignant clone could occur at the committed CD33+ myeloid pre- the modulation of inflammatory and immune responses through a cursor cell stage [59,60]. To test the assumption that CD33- precursors dampening effect on tyrosine kinase-driven signaling pathways [31, would be predominantly or completely normal in some of these cases, 32]. For example, in vitro studies have demonstrated that CD33 consti- CD33-depleted specimens from a small number of patients with such tutively suppresses the production of pro-inflammatory cytokines such leukemias were placed in long-term culture together with irradiated al- as IL-1β, TNF-α, and IL-8 by human monocytes in a sialic acid ligand- logeneic stroma cells; over time, CD33- precursors from some patients dependent and SOCS3-dependent manner [33]. Conversely, reduction indeed generated colony-forming cells with X chromosome inactivation of cell surface CD33 (e.g. via SOCS3 activity or RNA interference) or in- patterns consistent with predominantly non-clonal hematopoiesis [61, terruption of sialic acid binding can increase p38 mitogen-activated 62]. The demonstration of CD33 expression on AML-initiating cells in protein kinase (MAPK) activity and enhance cytokine secretion as well immunodeficient mice [63] has further been used to argue that CD33

• ANTIGENE DI DIFFERENZIAZIONE MIELOIDE ESPRESSO SUI NORMALI PRECURSORI MIELOIDI MULTIPOTENTI, UNIPOTENTI GRANULOCITI E MONOCITI, MA NON SULLE CELLULE STAMINALI EMOPOIETICHE

• PUO’ ANCHE ESSERE ESPRESSO SU ALCUNI SUBSET DI LINFOCITI B, T-ATTIVATI E CELLULE NK

• NON E’ ESPRESSO AL DI FUORI DEL SISTEMA EMATOPOIETICO

• LA DENSITA’ DI ESPRESSIONE DI SUPERFICIE E’ MOLTO VARIABILE (1-20 X 103 MOLECOLE/CELLULA) FUNZIONE PUTATIVA DI CD33 • ALCUNE EVIDENZE SPERIMENTALI SUGGERISCONO CHE LA MOLECOLA CD33 SIA COINVOLTA NELLA MODULAZIONE DELLE RISPOSTE INFIAMMATORIE E IMMUNI ATTRAVERSO UN EFFETTO INIBITORIO DI PATHWAYS DI SIGNALING CITOPLASMATICO TYROSINE-KINASE DRIVER

• STUDI IN VITRO HANNO DIMOSTRATO CHE CD33 SOPPRIME COSTITUTIVAMENTE LA PRODUZIONE DI CITOKINE PRO- INFIAMMATORIE COME IL-1β, TNF-α EIL-8 IN MANIERA DIPENDENTE DALL’ACIDO SIALICO E DA SOCS-3 (Sutherland D., Blood 2006; Orr S.J., Blood 2007; Lajaunias F., Eur J Immunol 2015) CD33

• CIRCA 85-90% DELLE AML SONO CD33-POSITIVE

• L’ESPRESSIONE DI CD33 E’ PIU’ ELEVATA E OMOGENEA NEI PROMIELOCITI LEUCEMICI

• ELEVATI LIVELLI DI CD33 SONO ASSOCIATI ALLA MUTAZIONE DI NPM1 E DI FLT3/ITD

• NEI PAZIENTI CON AML CD33 PUO’ ANCHE ESSERE EVIDENZIATA IN CIRCOLO COME MOLECOLA SOLUBILE. NON E’ CHIARO SE E IN CHE MISURA IL CD 33 SOLUBILE POSSA INTERFERIRE CON L’EFFICACIA TERAPEUTICA DEGLI ANTICORPI ANTI-CD33 A. Abdool et al./ Experimental Hematology 2010;38:462–471 467

Table 1. Performance of the quantitative immunoflow cytometry assay with cCD33 levels $570 U/uL also showed significantly developed for measuring plasma circulating CD33 shorter CRD (p 5 0.03) (Fig. 6). Median CRD was approx- Linear range, U/uL 750–12,000 imately 30 weeks for AML patients in the upper cCD33 Correlation factor, R2 (%) O99 quartile, compared to about 156 weeks in those with lower LOD, U/uL 642 levels (Fig 5). In MDS group, patients with cCD33 $570 LOQ, U/uL 750 U/uL had a median CRD of approximately 12 weeks, while Inter-assay variation (CV%) 3.2–8.4 patients with cCD33 !570 U/uL displayed median CRD of Intra-assay variation (CV%) 1.6–4.8 approximately 70 weeks (Fig. 6). CV 5 coefficient of variation; LOD 5 limit of detection; LOQ 5 limit of quantitation.

Discussion plasma cCD33 concentration can provide substantial predic- We reported here theCD33 development of a quantitative bead- tive/prognostic power for survival in ALL patients. based IFCExperimental assay for cCD33 measurement. Hematology To test 2010;38:462–471 this assay When we evaluated complete remission duration (CRD) for its diagnostic and prognostic utility, we analyzed base- in AML and MDS patients, Kaplan-Meier plots illustrated line cCD33 in three acute leukemia patient populations. that plasma cCD33 had substantive prognostic value in pre- The surface CD33 expression in acute leukemia (AML, dicting CRD for AMLCirculating and MDS. Higher cCD33 levels pre- CD33ALL, acute and promyelocytic its leukemia) clinical has already been value in acute leukemia dicted shorter CRD in both AML and MDS cases. In AML, documented in several studies [6–8,17,18]. Although it is cCD33 levels $1,500 U/uL werea correlated significantly considered myeloida marker, it is rarely expressed in ALL b b a withAdam shorter CRD Abdool (p 5 0.04) (Fig. 5,), Chen-Hsiung and MDS patients cases. Yeh Circulating, CD33 Hagop appears to Kantarjian be a marker capable , Susan O’Brien , JeanMarie Bruey , Francis Gilesc, and Maher Albitara A. Abdool et al./ Experimental Hematology 2010;38:462–471 469 aHematology and Oncology Department, Quest Diagnostics Nichols Institute, San Juan Capistrano, Calif., USA; bLeukemia, M.D. Anderson Cancer Center, University of Texas, Houston, Tex., USA; cUniversity of Texas, Health Science Center, San Antonio, Tex., USA (Received 26 October 2009; revised 23 February 2010; accepted 25 March 2010)

Objective. CD33 is a cell surface antigen for committed myelomonocytic lineage. We explored the potential of detecting CD33 as cell-free circulating protein in patients with leukemia. Materials and Methods. We developed a quantitative bead-based immunoflow cytometry assay to measure cell-free circulating CD33 (cCD33) levels in the plasma of patients with acute leukemia, and correlated these results with corresponding clinical behavior. We measured cCD33 levels in the plasma of 48 healthy subjects and in patients with acute myelogenous leukemia (n [ 98), acute lymphoblastic leukemia (n [ 46), myelodysplastic syndrome (MDS) (n [ 50), and myeloproliferative disorder (n [ 49). Results. Patients with acute myeloid leukemia and myeloproliferative disorders had signifi- cantly higher concentrations of cCD33 than the other patient groups and normal individuals (p [ 0.0001), and among these groups, MDS patients displayed the lowest cCD33 levels (p [ + 0.02). Circulating CD33 values correlated positivelyFigure 5. Kaplan-Meier with projection the CD33 of complete remissionblast duration cell (CRD) counts in 60 acute in myeloid these leukemia (AML) patients. CRD curves show clear separation patients. While there was no correlation betweenwith baseline cCD33 circulating CD33 levels (cCD33) and values survivalbelow (top or blue in line) acute and above the myelog- upper quartile, 1,500 U/uL (bottom or red line). The two curves are significantly different, showing shorter CRD in patients with higher levels of cCD33 (p 5 0.04, log-rank test). N 5 number of patients; E 5 patients with an enous leukemia and MDS, higher cCD33 plasmaevent, which concentrations is defined here as relapse. did correlate with shorter Figure 3. Plasma circulating CD33 levels in various leukemia patients. Levels of circulating CD33 (cCD33) in patients with acute myeloid leukemia (AML), acute lymphoblastic leukemiasurvival (ALL), myelodysplastic in acute syndrome (MDS), lymphoblastic myeloproliferative disorder leukemia (MPD), or normal ( subjectsp [ were0.03), measuredAbdool A. et al Exp. Hematology 2010 by and with shorter complete remission quantitative immunoflow cytometryduration and the box plots in with acute mean, mean myelogenous6 standard error (SE), and leukemia mean 6 1.96*SE were (p shown.[ Data0.04) were expressed and MDS (p [ 0.03). as cCD33 U/uL. Statistically significant difference was found between AML, MPD patients, and normal (p 5 0.0001), between MDS patients and normal (p 5 0.02). The number of patientsConclusion. in each group was alsoCirculating indicated in parentheses. CD33 can be detected in the plasma from patients with leukemias, and cCD33 levels may have clinical implication, e.g., predictive and prognostic value, in these patients. Ó 2010 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.

CD33 is a cell surface glycoprotein largely restricted to the studies indicate the involvement of Syk, c-Cbl, Vav, and myeloid/monocytic lineage. It binds sialic acids and there- ZAP-70 [4]. Alteration of CD33 signaling pathway might fore is a member of the Siglec (sialic acid binding also affect susceptibility to anti-CD33 therapy [5]. immunoglobulin-related lectins) family, and is also referred CD33 is found to be expressed in O90% of acute to as Siglec-3 [1]. Although CD33 can mediate sialic acid– myeloid leukemia (AML) blasts, whereas pluripotent hema- dependent cell interactions as a receptor protein, its ligand topoietic stem cells, lymphoid cells, and nonhematopoietic and function in myeloid cells has yet to be determined. cells normally do not express the CD33 antigen [6–8]. However, the two tyrosine residues in the cytoplasmic tail Antibody-mediated therapy for AML, therefore, focused Figure 6. Kaplan-Meier projection of complete remission duration (CRD) in 28 myelodysplastic syndrome (MDS) patients. CRD curves show clear sepa- of CD33 were shown to be phosphorylated upon receptor ration withon baseline CD33 circulating CD33 as (cCD33) an values attractive below (top or blue target. line) and above Promising the upper quartile, 570 clinical U/uL (bottom or red data line). The two curves are significantly different showing shorter CRD in patients with higher levels of cCD33 (p 5 0.03, log-rank test). N 5 number of patients; E 5 patients with crosslinking or by pharmacological treatment, e.g., perva- an event, whichhave is defined been here as relapse. shown with immunoconjugates consisting of nadate, and this, in turn, recruits two Src homology-2 the humanized anti-CD33 antibody (gemtuzumab) and the domain-containing tyrosine phosphatases, SHP-1 and cytotoxic drug calicheamicin (ozogamicin) as an effective SHP-2, to CD33 [2,3]. Downstream pathways in CD33 anti-AML regimen (Mylotarg CMA-676; Wyeth Pharma- signaling are not well-characterized, but some experimental ceuticals, Philadelphia, PA, USA) [9–12]. The potent inter- calating agent, ozogamicin, is released only at intracellular compartment (at lower pH) following a selective binding Offprint requests to: Maher Albitar, M.D., Quest Diagnostics Nichols to CD33-positive cells and receptor complex internalization Institute, 33608 Ortega Highway, San Juan Capistrano, CA 92675; (endocytosis), which in turn induces double-stranded DNA E-mail: [email protected] breaks, resulting in target cell death [13]. Gemtuzumab

0301-472X/$–see front matter. Copyright Ó 2010 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. doi: 10.1016/j.exphem.2010.03.016 26 D. Hoelzer, N. Gökbuget / Blood Reviews 26 (2012) 25–32

Table 1 The rationale for testing the addition of rituximab to chemotherapy in Expression of surface antigens for potential antibody therapy in ALL. ALL was an increase in overall survival by ≥20% in high grade Non- 7 Surface antigen ALL subtype Expression on Monoclonal Hodgkin Lymphoma studies. CD20 is found on about 30–40% of B- N20% of LBC antibody precursor ALL blasts, and its expression is more common in older adults Thiela Raponi4 with B-precursor ALL (40–50%). Furthermore CD20 is expressed in the majority of mature B-ALL blast cells (80–100%). CD19 B-precursor 95% 100% Blinatumomab Mature B-ALL 94% 100% CD20 B-precursor 41% 22–30% Rituximab Mature B-ALL 86% 100% 2.1. Prognostic impact of CD20 expression CD22 B-precursor 60–85% 93–96% Epratuzumab Mature B-ALL 69% 100% The prognostic impact of CD20 expression in B-precursor ALL is CD33 B-precursor 23% 17–26% Gemtuzumabb T-precursor 40% ozogamicin controversial (Table 2). In the largest series of 253 adult B-lineage 8 Ph+ ALL 9% ALL patients from MD Anderson Hospital, including 52 Ph+ ALL CD52 B-precursor 79% Alemtuzumab cases, the complete remission duration (CRD) at 3 years as well as T-precursor 77% the overall survival (OS) was inferior for CD20+ ALL with 20% and a Data from the German Multicentre Study Group for Adult ALL (GMALL) central Immu- 27% compared to a CRD of 55% and an overall survival of 40% for the nophenotyping, E. Thiel, S. Schwartz, Berlin, Germany (personal communication). CD20− ALL group. The outcome was particularly favorable for the b Not available anymore. CD20− younger age group. In contrast, another recent series of 143 15–60 years old patients showed no significant difference between Hoelzer D. et al Blood Reviews 2012 the 97 CD20− ALL patients with a complete remission (CR) rate, such as profound B- or T-cell lymphopenia with related clinical conse- event free survival (EFS) and OS of 92%, 54% and 59% compared to quences, particularly infections. the 46 CD20+ ALL cases with 89%, 47%, and 55%.9 However there was a negative prognostic value of CD20 expression within the high 9 2. CD20 antigen risk patients (WBC≥30×10 /L) due to a higher cumulative incidence of relapse (Table 2). Similarly in a study with 169 children,10 CD20 CD20 is a 33–37 kDa non-glycosylated B-lymphocyte specificintegral expression was not associated with inferior outcome; EFS 84% in membrane phosphoprotein. The function of CD20 has not been clarified CD20+ vs. 78% for CD20− ALL. The most recent results from Bacha- in detail but it seems to be involved in the regulation of transmembrane nova and colleagues suggest that an adverse prognostic impact of calcium conductance.5 CD20 is only very slowly internalized. CD20 expression in ALL can be overcome by allogeneic hematopoietic CD20 is expressed on normal and malignant B lymphocytes, but not stem cell transplantation. They analyzed a total of 125 patients with on normal stem cells. Rituximab is a chimeric human/mouse monoclonal pre-B-ALL after SCT and investigated both children and adults. The antibody against CD20. From in vitro studies, various mechanisms of ac- overall 5 year DFS was 43% in CD20− and 55% in CD20+ ALL. Also tion have been postulated, including complement-dependent cytotoxici- the individual DFS results for children with 50% in CD20− ALL as ty, antibody-dependent cellular cytotoxicity, and induction of apoptosis.6 compared to 62% for CD20+ ALL and in adults with 40% vs. 48%

Table 2 Prognostic impact of CD20 expression in B-lineage acute lymphoblastic leukemia.

CD20 Neg. CD20 Pos.

Jeha et al. N=169 Blood 2006 108; 3302–3304 Age Children b1,b10,N10 years EFS 78% 84% n.s. Thomas et al. N=253 (52 Ph+) Blood 2009 113, 6330–6337 Age 38 (15–80) 39 (16–80) Relapse rate 33% 68% intensified Hyper CVAD 35% 80% conventional VAD/CVAD p=0.01 CRD 3 years 55%2 20% OS 40%2 27% Maury et al. N=143 Haematologica 2010, 324–328 N 97 46 Age 36 (15–60) 33 (17–60) n.s. CR 92% 89% n.s. EFS 54% 47% n.s. OS 59% 55% n.s. HR: WBCN30×109/L Relapse rate 24% 70% p=0.006 EFS (42 months) 59% 15% p=0.003 Bachanova et al. N=125 after allogeneic SCT Blood 2011, 117: 5261–5263 N 67 58 Overall Age 27 (0.6–66) n.s. DFS 5 years 43% 55% n.s. 2–4 acute GvHD 43% 38% n.s. TRM 19% 18% n.s. Children: N=54 48% 52% n.s. DFS 5 years 50% 62% n.s. Relapse 3 years 38% 17% n.s. Adults: N=71 61% 39% n.s. DFS 5 years 40% 48% n.s. Relapse 3 years 29% 25% n.s. From www.bloodjournal.org by guest on May 9, 2018. For personal use only.

BLOOD, 15 DECEMBER 2005 ⅐ VOLUME 106, NUMBER 13 HSCs EXPRESS MYELOID MARKERS 4089

Table 3. Characteristics of AML patient samples WBC count, CD33, % of CD33, % of Patient no. ؋ 109/L FAB group Karyotype MNCs CD34؉CD38؊,% CD34؉CD38؊

1 151 1 Normal 76 0.24 19 2 40 2 Normal 68 0.37 26 From www.bloodjournal.org by guest on May 9, 2018. For personal use only. 3* 22 2 ϩ11, ϩ13 96 13.5 100 4* 66 2 t(8;21) 98 61 100 BLOOD, 15 DECEMBER 2005 ⅐ VOLUME 106, NUMBER 13 HSCs EXPRESS MYELOID MARKERS 4087 5 73 2 Normal 94 0.68 76 6 2.5 4 Normal 43 0.04 26 7714ϩ3, ϩ10 88 10 100 United Kingdom) and the University of Pennsylvania School of Medicine Effect of monoclonal antibodies on engraftment 8 115 5 Normal 87 0.04 23 (Philadelphia, PA), after informed consent was provided. We obtained cord 9† 12 NA t(11;19) 58 0.47 95 Cells were incubated for 30 minutes at 4°C with either the test antibody or From www.bloodjournal.org by guest on May 9, 2018. For personal use only. blood from mothers attending University College Hospital (London, United 10† 19 NA Complex 79 31 96 isotype control. The cells were washed prior to injection into irradiated, Kingdom) and bone marrow from healthy volunteer donors at the Univer- 11 2.2 4 Normal 96 7.8 83 age- and sex-matched mice. Accessory cells were added after the wash step sity of Pennsylvania, after informed consent was provided. The protocol 12 77 4 Inv16 87 10.8 96 for cord blood LinϪCD34ϩCD38Ϫ cells. Mice were killed at 6 or 12 weeksHEMATOPOIESIS was approved by the hospital research ethics committees (University of CD33 PUO’ 13ESSERE33 ESPRESSA5 Normal ANCHE47 A0.15 14 Pennsylvania Review Board and the East London Ethical Committee). and the bone marrow was assessed for engraftment as described (see From www.bloodjournal.org by guest on May 9, 2018. For personal use only. “Analysis of murine bone marrow”). WBC indicates white blood cell; FAB, French-American-British classification; NA, not applicable. Mononuclear cells (MNCs) were obtained by Ficoll-Paque density *Samples taken at relapse. centrifugation. LIVELLOHEMATOPOIESIS †Therapy-relatedDELLE AML. CELLULE STAMINALI Serial transplantation Hematopoietic stem cells express multiple myeloid markers: implications Mice LEUCEMICHEϪ Ϫ ϩ Ϫ Ϫ/Ϫ Mice were humanely killed 6 or 12 weeks after transplantation and bonerepresenting the CD33 fraction of 40 000 Lin CD34 CD38 injected into NOD/SCID-␤2m mice (Table 1). The frequency of ϩ ϩ All animal experiments were performed in compliance with Home Office marrow was obtained. Cells were stained with FITC-conjugatedfor anti–cells/mouse) the origin and engraftment and targeted was assessed therapy at 8 weeks. of Only acute 2 myeloidSRCs at 6 weeks leukemia was significantly higher in the CD13 CD123 (1 Hematopoietic stem cells express multiple myeloid markers:Ϫ ϩ implications and Cancer Research UK (CRUK) guidelines. NOD/SCID mice and HLA-A, -B, and -C and PE-conjugated anti–mouse CD45 antibodies (BDmice showed multilineage engraftment (Figure 2A). We concluded in 575 cells) and CD13 CD123 (1 in 1157 cells) fractions than in ϩ ϩ ϩ Ϫ Ϫ/Ϫ Biosciences) and DAPI. HLA-A , -B , -C mouse CD45 cellsDavid werethat C. the Taussig, majority Daniel ofJ. the Pearce, engrafting Catherine potential Simpson, is present Ama Z. inRohatiner, the T. AndrewϪ Lister, GavinϪ Kelly, Jennifer L. Luongo, NOD/SCID/␤2-microglobulin null (NOD/SCID-␤2m ) mice were origi- for the origin and targeted therapy of acutethe CD13 myeloidCD123 fraction leukemia (1 in 4383 cells; P Ͻ .05). Indeed, as nally obtained from Dr Leonard Schultz (Jackson Laboratory, Bar Harbor, Gwenn-ae¨lCD33ϩ fraction. H. Danet-Desnoyers, and Dominique Bonnet few as 300 CD13ϩCD123ϩ cells could engraft. ME) and bred at Charles Rivers Laboratories (Margate, United Kingdom). DavidThe C. CD33 Taussig,ϩ fraction Daniel was J. shownPearce, to Catherine be capableSimpson, of self-renewal Ama by Z. Rohatiner,Multilineage T. Andrew engraftment Lister, Gavin was Kelly, also seen Jennifer in the L. moreLuongo, stringent They were kept in microisolators and fed sterile food and acidified water. Gwenn-ae¨lperforming H. serial Danet-Desnoyers, transplantation. and Bone Dominique marrow cells Bonnet were har- NOD/SCID mouse at 6 weeks from all 4 fractions (Table 2). Mice aged 8 to 12 weeks were irradiated at 375 rads (137Cs source) up to 24 vested from mice that had successfully received transplants of cells Together, these data provide further evidence for the presence Human hematopoietic stem cells (HSCs) human long-term repopulating cells from more, based on the phenotypic similarity hours before intravenous injection of cells. from the CD33ϩ fraction. Human cells (0.9-2 ϫ 106) were injected of myeloid markers on the surface of normal long-term repopulat- areHumaninto generally 7 secondary hematopoietic regarded recipients. as being stem Four devoid cells of these (HSCs) of 7cord secondary bloodhuman and recipients long-term bone marrow.ing repopulating cells. In addition, cellsof from HSCs andmore, LSCs, based our on data the provide phenotypic sup- similarity arewere generally engrafted regarded by human as cells. being devoid of cord blood and bone marrow. In addition, of HSCs and LSCs, our data provide sup- Immunophenotyping the markers expressed by differentiated we demonstrate that nonobese diabetic/ port for the hypothesis that AML derives bloodtheSimilar markers cells, the experiments lineage-specificexpressed were by performed differentiated antigens. usingsevere bonewe marrow demonstratecombined from immunodeficiency that nonobese diabetic/from an HSC.port Our forthe findings hypothesis also provide that AMLa derives Cells were stained with fluorescein isothiocyanate (FITC)–conjugated bloodhealthy cells, adults the (Figure lineage-specific 2B). The SRC antigens. frequency wassevere similar in combined each AML immunodeficiencyLSCs express CD33 andfrom CD13 an HSC. Our findings also provide a However, recent work suggests that genes (NOD/SCID)Ϫ ϩ leukemia-Ϫ initiating cells (SL- challenge to contemporary attempts to anti-CD38 and lineage cocktail 1 (CD3, CD14, CD16, CD19, CD20, However,fraction, but recent given work that suggests90% (Ϯ 11%) that of genes Lin CD34(NOD/SCID)CD38 cells leukemia- initiating cells (SL- challenge to contemporaryϩ Ϫ attempts to -The expression؉ of myeloid markers on the CD34 CD38 popula -CD56), peridinin chlorophyll protein complex (PerCP)–conjugated CD34, associatedassociatedare CD33ϩ withthe with majority the the myeloid myeloid of bone lineage marrow lineage are SRCs areICs) express areICs) restricted CD33are restrictedϩ. to the CD33 to thefraction CD33؉ fraction in improve in improve the outcome the of outcome AML using of AML my- using my phycoerythrin (PE)–conjugated anti-CD33 or anti-CD123, or allophycocya- tion of AML that is known to contain the SCID-leukemia initiating transcribedtranscribed in in mouse mouse HSCs. HSCs. Here, Here, we we11 of11 12 of acute 12 acute myeloid myeloid leukemia leukemia (AML)16 (AML)eloid antigen-targetedeloid antigen-targeted therapies, given therapies, the given the nin (APC)–conjugated CD13 (all antibodies from Becton Dickinson [BD] cells (SL-ICs) was studied. The majority of these cells expressed exploreMany SRCs whether express myeloid other myeloid genes are markers actu- samples studied, indicating that leukemic potential for HSC killing. (Blood. 2005; Biosciences, Oxford, United Kingdom). Analyses were performed on either explore whether myeloid genes are actu- samples studied, indicatingCD13, that CD33, leukemic and CD123potential (Figure for 1A,D). HSC MNCs killing. from (Blood. patients 2005; with ally translated in human HSCs. We show stem cells (LSCs) express this antigen. 106:4086-4092)ϩ Ϫ a BD Life Science Research (LSR) flow cytometer or a BD fluorescence allyWe translated investigated in humanwhether HSCs. cells expressing We show 2 otherstem myeloid cells (LSCs) markers, expressAML this (Table antigen. 3) were106:4086-4092) sorted into CD33 and CD33 fractions and that CD33, CD13, and CD123, well-estab- This study changes our view of HSCs and activated cell sorting (FACS) Aria flow cytometer after cells were thatCD13 CD33, and CD13, CD123, and were CD123, able well-estab- to repopulateThis the study bone marrowchanges of our viewinjected of HSCs into immunodeficient and mice (Figure 3A-D). The purity was resuspended in a DAPI (4,6 diamidino-2-phenylindole)–containing solu- lishedimmunodeficient myeloid markers, mice. Cord are blood expressed LinϪCD34 on ϩCD38theϪ processcells were of differentiation.98.8% (Ϯ 1.6%) and Further- 97.3% (Ϯ© 1.9%)2005 by for The the American CD33ϩ and Society CD33 ofϪ Hematology lished myeloid markers, are expressed on the process of differentiation. Further- © 2005 by The American Society of HematologyϪ/Ϫ tion. Gates were set up to exclude nonviable cells and debris. The negative Introductionsorted on the basis of CD13 and CD123 into 4 fractions and fractions, respectively. NOD/SCID or NOD/SCID-␤2m mice fraction was determined using appropriate isotype controls. Introduction Hematopoietic stem cells (HSCs) are generally regarded as being by GO. Furthermore, although GO may induce remission when Analysis of murine bone marrow Hematopoieticdevoid of lineage-specific stem cells (HSCs) antigens. are generally1-3 As HSCs regarded commit as being to specificby GO.given Furthermore, as a sole although agent, many GO may patients induce have remission relapses. when12 This may be blood cell lineages, lineage markers are expressed. However, it has because the leukemic stem cells (LSCs) are resistant to the toxin to Murine marrows were stained with human-specific FITC-conjugated anti- devoid of lineage-specific antigens.1-3 As HSCs commit to specific given as a sole agent, many patients have relapses.12 This may be been noted that genes associated with specific lineages are ex- which the antibody is conjugated.7,14 An alternative explanation CD19, PE-conjugated anti-CD33, and phycoerythrin-cyanin 5 (PE-Cy5)– bloodpressed cell lineages, in cells lineage with a markers stem arecell expressed. phenotype. However,4 This it led has tobecause the is the that leukemic LSCs, stem unlike cells (LSCs) the majority are resistant of leukemic to the toxin blasts, to do not conjugated anti-CD45 antibodies, and DAPI. A BD LSR flow cytometer 7,14 was used for analysis. More than 100 000 events were collected. Engraft- beendevelopment noted that of genes the “lineage associated priming” with specific hypothesis lineages that are proposes ex- thatwhich theexpress antibody CD33. is conjugated. An alternative explanation ment of AML was said to be present if a single population of pressedHSCs promiscuouslyin cells with a express stem cell lineage-specific phenotype.4 This genes led prior to the to com-is that LSCs,In this unlike study, the nonobese majority diabetic/severe of leukemic blasts, combined do not immunodefi- CD45ϩCD33ϩCD19Ϫ cells was present. Normal multilineage engraftment developmentmitment.5 Using of the “lineagea conditional priming” knockout hypothesis mouse, that proposesone group that recentlyexpressciency CD33. (NOD/SCID) mice were used to phenotype human HSCs. was defined by the presence of separate CD45ϩCD33ϩ and CD45ϩCD19ϩ We show that HSCs express a number of myeloid markers HSCsdemonstrated promiscuously that express the myeloid lineage-specific gene lysozyme genes prior is transcribed to com- inIn this study, nonobese diabetic/severe combined immunodefi- populations with appropriate scatter characteristics. HSCs.6 This appeared to confirm the “lineage priming” hypothesis. including CD13, CD33, and CD123. We also show that AML LSCs 5 ciency (NOD/SCID) mice were used to phenotype human HSCs.15,16 mitment.We decidedUsing a conditional to take the knockout investigation mouse, a one step group further recently and asked express CD33 and CD13 using similar methodology. The data We show that HSCs express a number of myeloid markers Assessment of engraftment potential of AML demonstratedwhether markers that the previously myeloid thought gene lysozyme to be restricted is transcribed to the inmyeloid overturn the dogma that human HSCs are devoid of myeloid HSCs.lineage6 This are appeared actually to expressed confirm the by “lineage human priming” HSCs. hypothesis. This questionincluding is markers CD13, CD33,and change and CD123. our view We also of showdifferentiation. that AML LSCs This provides a Samples were screened to assess whether they had the potential to engraft also of considerable clinical importance; a number of therapies thatexpresschallenge CD33 and toCD13 contemporary using similar attempts methodology. to improve15,16 The datathe outcome of NOD/SCID and NOD/SCID-␤ mϪ/Ϫ mice. MNCs (107) were injected into We decided to take the investigation a step further and asked 2 target cells expressing myeloid markers are under development for AML using myeloid antigen-targeted therapies given the potential each mouse. Engraftment of AML was confirmed, where possible, with whether markers previously thought to be restricted to the myeloid overturn the dogma that human HSCs are devoid of myeloid the treatment of acute myeloid leukemia (AML).7-10 These thera- for HSC killing. Furthermore, our data indicate that the phenotypes morphology, phenotyping, and fluorescence in situ hybridization. Samples lineage are actually expressed by human HSCs. This question is markers and change our view of differentiation. This provides a Ϫ/Ϫ piesFigure were 3. CD33 designed sorting strategy with for the AML aim and ofengraftment selectively time course killing of sorted leukemic fractions.ofAML LSCs MNCs andwere staineda population with anti-CD33 of HSCs antibody are and similar, sorted into consistent CD33Ϫ and with the that showed significantly higher engraftment in NOD/SCID-␤2m mice 17 ϩ 11 challengehypothesisTaussig D.C. et al Blood 2005 to contemporary that AML attempts arises from to improve an HSC. the outcomeϪ of ϩ were injected into this strain in sorting experiments. Figure 1. Phenotype of cord blood, bone marrow, and AML cells. Flow cytometryalsoblastsCD33 of considerablefractions. that express (A) AML clinical myeloid cells stained importance; antigens with isotype asuch control.number as (B) CD33, of AML therapies cellswhile stained that with sparing anti-CD33 antibody and sort gates G1 and G2. Purity of the sorted CD33 (C) and CD33 was used to assess the phenotype of cord blood, bone marrow, and AML cells.(D) (A) fractions of AML. Engraftment of CD33ϩ and CD33Ϫ fractions of AML samples no. 1AML (E) and using no. 9 (F) myeloid at 6 and antigen-targeted 12 weeks. Filled squares therapies represent given mice giventhe potential CD33Ϫ cells. normal HSCs. If HSCs also expressϩ myeloid markers, HSCs will be Percentage of (LinϪ) CD34ϩCD38Ϫ cells expressing CD13, CD33, and CD123targetOpen in cells squares expressing represent mice myeloid given CD33 markerscells. are under development for bone marrow (n 6), cord blood (n 12), and AML (n 18). The error barstargeted along with leukemic cells. One specific7-10 therapy, gemtu-for HSC killing. Furthermore, our data indicate that the phenotypes FACS analysis ϭ ϭ ϭ the treatment of acute myeloid leukemia (AML). These thera- represent SD. In panels B-D, the expression profile of CD34 and CD38 is shownzumab in ozogamicin (GO), an antibody against CD33 thatof is LSCsMaterials and a population and of HSCsmethods are similar, consistent with the the left plot. (LinϪ) CD34ϩCD38Ϫ cells are denoted by the R1 gate for further analysis.pies were designed with the aim of selectively killing leukemic We depleted cord blood, bone marrow, and AML samples no. 11 and no. 12 conjugated to a cytotoxic agent,7,12,13 is undergoing clinical trials in 17 The expression of CD33 and CD13/CD123 on cells in the R1 gate is shownblasts in the that express myeloid antigens such as CD33,11 while sparing hypothesisPrimary that AML cells arises from an HSC. of cells expressing lineage markers using StemSep columns and human middle and right plots, respectively. Cells stained with isotype control are representedthousands of patients in the United States and Europe. Prolonged progenitor enrichment cocktail (StemCell Technologies, Vancouver, BC, by the open histogram in the middle plot. More than 98% of cells stained with isotypenormalcytopenias HSCs. have If HSCs been also noted express despite myeloid successful markers, clearance HSCs will of be leuke- Peripheral blood cells were obtained from adult patients with newly Canada) prior to sorting. The sorting was performed either on a MoFlo cell control fall within the bottom left quadrant in the right plot. (B) Expression profile of 12,13 Ϫ ϩ Ϫ targetedmic cells along in patients with leukemic receiving cells. GO. One specificThis may therapy, reflect gemtu- HSC killing diagnosed and relapsed AML at St Bartholomew’s Hospital (London, sorter (DakoCytomation Colorado, Fort Collins, CO) or a BD FACS Aria. CD33, CD13, and CD123 on Lin CD34 CD38 cord blood. (C) Expression profile of CD33, CD13, and CD123 on LinϪCD34ϩCD38Ϫ bone marrow. (D) Expression profile Gates were set up to exclude nonviable cells and debris. The purity of sorted zumab ozogamicin (GO), an antibody against CD33 that is Materials and methods of CD33, CD13, and CD123 on CD34ϩCD38Ϫ AML cells (sample no. 10). (E) Mean 7,12,13 fractions was assessed to ensure sort quality. We mixed sorted cells with Ϫ ϩ Ϫ fluorescence intensity of antigens on (Lin ) CD34 CD38 cells from bone marrow,conjugatedFrom the Haematopoietic to a cytotoxic Stemagent, Cell Laboratory,is undergoing Cancer clinical Research trials UK, in London Institutes of Health (D.B.). D.C.T. and D.J.P. contributed equally to this study. 500 000 irradiated (1500 rads) accessory cells prior to injection into mice. cord blood, and AML. Research Institute, London, United Kingdom; Division of HaematologicalPrimary cells thousands of patients in the United States and Europe. Prolonged The online version of the article contains a data supplement. cytopeniasOncology, have St Bartholomew’s been noted despite Hospital, successful West Smithfield,clearance of London, leuke- United Kingdom; Fluorescence Activated Cell Sorting Laboratory, Cancer ResearchPeripheralAn bloodInside cellsBlood wereanalysis obtained of this from article adult appears patients at the with front newly of this issue. micUK, cells London in patients Research receiving Institute, GO. London,12,13 This United may reflect Kingdom; HSC killingComputationaldiagnosed and relapsed AML at St Bartholomew’s Hospital (London, Reprints: Dominique Bonnet, Rm 504, London Research Institute, 44 Lincoln’s Genome Analysis Laboratory, Cancer Research UK, London Research Inn Fields, London, WC2A 3PX, United Kingdom; e-mail: d.bonnet Institute, London, United Kingdom; and Department of Hematology/Oncology, @cancer.org.uk. University of Pennsylvania School of Medicine, Philadelphia, PA. From the Haematopoietic Stem Cell Laboratory, Cancer Research UK, London InstitutesThe of Health publication (D.B.). D.C.T. costs and of this D.J.P. article contributed were equally defrayed to this inpart study. by page charge ResearchSubmitted Institute, March London,16, 2005; United accepted Kingdom; August Division 3, 2005. of Prepublished Haematological online as payment. Therefore, and solely to indicate this fact, this article is hereby Blood First Edition Paper, August 30, 2005; DOI 10.1182/blood-2005-03-1072.The online version of the article contains a data supplement. Oncology, St Bartholomew’s Hospital, West Smithfield, London, United marked ‘‘advertisement’’ in accordance with 18 U.S.C. section 1734. Kingdom;Supported Fluorescence by Cancer Research Activated UKCell and Sorting by a Laboratory,grant (HL64856-01) Cancer Research from the NationalAn Inside©Blood 2005analysis by The ofAmerican this article Society appears of at Hematology the front of this issue. UK, London Research Institute, London, United Kingdom; Computational Reprints: Dominique Bonnet, Rm 504, London Research Institute, 44 Lincoln’s Genome Analysis Laboratory, Cancer Research UK, London Research Inn Fields, London, WC2A 3PX, United Kingdom; e-mail: d.bonnet Institute, London, United Kingdom; and Department of Hematology/Oncology, 4086 @cancer.org.uk. BLOOD, 15 DECEMBER 2005 ⅐ VOLUME 106, NUMBER 13 University of Pennsylvania School of Medicine, Philadelphia, PA. The publication costs of this article were defrayed in part by page charge Submitted March 16, 2005; accepted August 3, 2005. Prepublished online as payment. Therefore, and solely to indicate this fact, this article is hereby Blood First Edition Paper, August 30, 2005; DOI 10.1182/blood-2005-03-1072. marked ‘‘advertisement’’ in accordance with 18 U.S.C. section 1734. Supported by Cancer Research UK and by a grant (HL64856-01) from the National © 2005 by The American Society of Hematology

4086 BLOOD, 15 DECEMBER 2005 ⅐ VOLUME 106, NUMBER 13 UNCONJUGATED ANTIBODIES

MoAb Characteriscs Clinical Results

HUMANIZED LINTUZUMAB VERY MODEST IgG1 (SGN-33) ACTIVITY AS SINGLE AGENT. FAILED TO CLINICAL IMPROVE DEVELOPMENT SURVIVAL WHEN ABANDONED ADDED TO MAb 33.1/ FULLY HUMAN CONVENTIONAL B1 836858 IgG1 ENGINEERED CHEMOTHERAPEU TO HAVE TICS INCREASED ADCC 146 G.S. Laszlo et al. / Blood Reviews 28 (2014) 143–153 146 G.S. Laszlo et al. / Blood Reviews 28 (2014) 143–153

Fig. 2. Schematic structure of GO. The humanized IgG4 CD33 antibody is conjugated to the calicheamicin-γ1 derivative via a hybrid 4-(4′-acetylphenoxy)butanoic acid linker. GO has Fig. 2. Schematic structure of GO. The humanized IgG CD33 antibody is conjugated to the calicheamicin-γ derivative via a hybrid 4-(4′-acetylphenoxy)butanoic acid linker. GO has approximately 50% of the antibody loaded with 4–6 mol of the toxic4 moiety per mole of antibody; the remaining1 50% of the antibody molecules are unconjugated. approximately 50% of the antibody loaded with 4–6 mol of the toxic moiety per mole of antibody; the remaining 50% of the antibody molecules are unconjugated. Reprinted from Current Opinion in Pharmacology [165] with permission from Elsevier. Reprinted from Current Opinion in Pharmacology [165] with permission from Elsevier. stranded breaks. This DNA damage elicits a strong cellular response Laszlo G.S. et al Blood Reviews 2014, 143-153 drug efflux, but these as well as other studies combining GO with stranded breaks. This DNA damage elicits a strong cellular response drug efflux, but these as well as other studies combining GO with with cell cycle arrest followed by either DNA repair or, if damage is over- conventional therapeutics or alternative types of chemosensitizers with cell cycle arrest followed by either DNA repair or, if damage is over- conventional therapeutics or alternative types of chemosensitizers whelming, apoptosis and cell death, predominantly through mitochon- have been hampered by small sample sizes and absence of control whelming, apoptosis and cell death, predominantly through mitochon- have been hampered by small sample sizes and absence of control drial pathways with cytochrome c release, involvement of Bcl-2 family groups, and have produced mixed results that are difficult to inter- drial pathways with cytochrome c release, involvement of Bcl-2 family groups, and have produced mixed results that are difficult to inter- proteins, and caspase activation [58,81,82]. pret [86–91,99]. proteins, and caspase activation [58,81,82]. pret [86–91,99]. GO was given accelerated U.S. marketing approval in 2000 for treat- GO was given accelerated U.S. marketing approval in 2000 for treat- ment of adults N60 years of age with CD33+ AML in first relapse who 4.2.1.3. Randomized trials of GO as add-on to intensive AML induction ment of adults N60 years of age with CD33+ AML in first relapse who 4.2.1.3. Randomized trials of GO as add-on to intensive AML induction were not candidates for standard cytotoxic chemotherapy. Approval chemotherapy. A clearer picture has emerged from 5 large, randomized were not candidates for standard cytotoxic chemotherapy. Approval chemotherapy. A clearer picture has emerged from 5 large, randomized was based on interim data from 3 open-label, multicenter single-arm studies conducted in Europe and the U.S. that have investigated GO was based on interim data from 3 open-label, multicenter single-arm studies conducted in Europe and the U.S. that have investigated GO trials showing an overall response rate (complete remission [CR] and as an addition to conventional chemotherapy in adults with newly trials showing an overall response rate (complete remission [CR] and as an addition to conventional chemotherapy in adults with newly CR with incomplete platelet count recovery [CRp]) of almost 30% in diagnosed AML. In 4 of the studies (MRC/NCRI AML15 and AML16, CR with incomplete platelet count recovery [CRp]) of almost 30% in diagnosed AML. In 4 of the studies (MRC/NCRI AML15 and AML16, 142 adults with CD33+ de novo AML in first relapse who typically ALFA0701, and GOELAMS AML 2006 IR), the use of GO during induction 142 adults with CD33+ de novo AML in first relapse who typically ALFA0701, and GOELAMS AML 2006 IR), the use of GO during induction received two 9 mg/m2 doses of GO 14 days apart [3,83]; this dose was resulted in statistically significantly improved survival in similar subsets received two 9 mg/m2 doses of GO 14 days apart [3,83]; this dose was resulted in statistically significantly improved survival in similar subsets chosen because it provided complete saturation of CD33 binding sites of patients [100–103], particularly those with “favorable” prognosis chosen because it provided complete saturation of CD33 binding sites of patients [100–103], particularly those with “favorable” prognosis without dose-limiting non-hematologic toxicities in an earlier phase 1 AML (as defined by core-binding factor translocations) or those with without dose-limiting non-hematologic toxicities in an earlier phase 1 AML (as defined by core-binding factor translocations) or those with study [84]. The final report on 277 patients confirmed the early results normal cytogenetics, the majority of whom have been reproducibly study [84]. The final report on 277 patients confirmed the early results normal cytogenetics, the majority of whom have been reproducibly in this patient population with an overall response rate of approximately shown to live longer if given GO. GO also improved outcome when com- in this patient population with an overall response rate of approximately shown to live longer if given GO. GO also improved outcome when com- 25%, although remission durations were relatively short [85]. bined with FLAG-Ida [100], suggesting that GO was not acting merely as 25%, although remission durations were relatively short [85]. bined with FLAG-Ida [100], suggesting that GO was not acting merely as a non-specific means of intensification of standard-dose cytarabine- a non-specific means of intensification of standard-dose cytarabine- 4.2.1.2. Phase 2 experience with GO. Subsequently, several phase 2 studies containing induction therapy. There was no effect of GO on survival in 4.2.1.2. Phase 2 experience with GO. Subsequently, several phase 2 studies containing induction therapy. There was no effect of GO on survival in investigated GO in unselected patients with newly diagnosed and/or the 5th trial, SWOG S0106 [104]. Unlike the European trials, however, investigated GO in unselected patients with newly diagnosed and/or the 5th trial, SWOG S0106 [104]. Unlike the European trials, however, relapsed/refractory non-APL AML. While they confirmed single agent in which identical doses of conventional chemotherapeutics were relapsed/refractory non-APL AML. While they confirmed single agent in which identical doses of conventional chemotherapeutics were activity of GO, the overall response rates have usually not exceeded used in the +/− GO arms, S0106 used a lower anthracycline dose activity of GO, the overall response rates have usually not exceeded used in the +/ GO arms, S0106 used a lower anthracycline dose 25–35% and were occasionally quite disappointing, particularly in with GO than− the control arm, possibly accounting for some of the 25–35% and were occasionally quite disappointing, particularly in with GO than the control arm, possibly accounting for some of the heavily pretreated patients [86–91]. Of note, in most of the non-APL differences in outcomes between these trials. Furthermore, even in heavily pretreated patients [86–91]. Of note, in most of the non-APL differences in outcomes between these trials. Furthermore, even in AML studies, GO was given at 9 mg/m2 every 2 weeks. However, new S0106 longer survival was seen in patients with core-binding factor AML studies, GO was given at 9 mg/m2 every 2 weeks. However, new S0106 longer survival was seen in patients with core-binding factor CD33 binding sites continuously arise and surface CD33 levels return AML. A recent individual patient meta-analysis of these 5 randomized CD33 binding sites continuously arise and surface CD33 levels return AML. A recent individual patient meta-analysis of these 5 randomized to pretreatment levels within 72 h after antibody administration [40, trials indicated that the addition of GO significantly reduced the risk of to pretreatment levels within 72 h after antibody administration [40, trials indicated that the addition of GO significantly reduced the risk of 74]. Hence, repeated administration of lower, (near-) saturating doses relapse (hazard ratio [HR] = 0.80 [95% confidence interval, 0.72–0.89], 74]. Hence, repeated administration of lower, (near-) saturating doses relapse (hazard ratio [HR] = 0.80 [95% confidence interval, 0.72–0.89], of GO every 3 days may enhance intracellular drug accumulation. The P = 0.00006), leading to improved relapse-free survival (HR = 0.84 of GO every 3 days may enhance intracellular drug accumulation. The P = 0.00006), leading to improved relapse-free survival (HR = 0.84 Acute Leukemia French Association (ALFA) group used such a fraction- [0.76–0.94], P = 0.001) and overall survival (HR = 0.89 [0.82–0.97], Acute Leukemia French Association (ALFA) group used such a fraction- [0.76–0.94], P = 0.001) and overall survival (HR = 0.89 [0.82–0.97], ated dosing schedule and showed it to have promising efficacy and P = 0.01) despite a slightly greater early mortality (P = 0.08). In con- ated dosing schedule and showed it to have promising efficacy and P = 0.01) despite a slightly greater early mortality (P = 0.08). In con- acceptable toxicity [92,93], although no direct comparisons have been trast to the improvement in survival, the addition of GO did not change acceptable toxicity [92,93], although no direct comparisons have been trast to the improvement in survival, the addition of GO did not change conducted to demonstrate superiority over the traditional administra- the remission rates during induction. As suggested by the individual conducted to demonstrate superiority over the traditional administra- the remission rates during induction. As suggested by the individual tion schedule. In contrast to its limited effectiveness in non-APL AML, studies, there was a highly significant interaction between the treat- tion schedule. In contrast to its limited effectiveness in non-APL AML, studies, there was a highly significant interaction between the treat- GO + all-trans retinoic acid (ATRA) produced CR and relapse-free ment effect and cytogenetic risk group, with the benefit of GO being GO + all-trans retinoic acid (ATRA) produced CR and relapse-free ment effect and cytogenetic risk group, with the benefit of GO being survival rates similar to those seen with an anthracycline + ATRA in primarily seen in patients with favorable-risk disease (HR = 0.50 survival rates similar to those seen with an anthracycline + ATRA in primarily seen in patients with favorable-risk disease (HR = 0.50 newly diagnosed APL [94,95], while single agent GO was found to [0.33–0.77], P = 0.001) and, to a lesser degree, intermediate-risk – P routinely eliminate evidencenewly diagnosed of APL APL in[94,95] patients, while with single molecular agent GO was relapse found to [0.33disease0.77], (HR= 0.001) = and, 0.85 to [0.76a lesser– degree,0.96], intermediate-riskP = 0.007) but not those with – P of APL [96–98]. The distinctiveroutinely eliminate sensitivity evidence of of APL APL in cells patients to withGO likely molecular results relapse diseaseadverse-risk (HR = 0.85 disease [0.76 0.96], (HR == 1.04 0.007) [0.86 but not–1.25], thoseP with= 0.7) [105]. These find- – – P fi from the disease's CD33-richof APL [96 98] nature. The distinctive and lack sensitivity of signi of APLficant cells to drug GO likely trans- results adverse-riskings in adultdisease (HR patients = 1.04 [0.86 are1.25], complemented= 0.7) [105]. These by recentnd- data from a large fi porter activity. from the disease's CD33-rich nature and lack of signi cant drug trans- ingsrandomized in adult patients pediatric are complemented trial (COG-AAML0531) by recent data from a large in over 1000 individuals Because GO is subjectporter activity. to drug pump-mediated extrusion from randomizedb 30 years pediatric of age, trial in(COG-AAML0531) whom the addition in over 1000 of individuals GO to conventional intensive b cells, several studiesBecause have combined GO is subject to GO drug with pump-mediated agents extrusionthat block from 30chemotherapy years of age, in whom was the associated addition of GO with to conventional a signi intensiveficantly improved event-free cells, several studies have combined GO with agents that block chemotherapy was associated with a significantly improved event-free MYLOTARG: Mechanism of acon

MYLOTARG Extracelluar space 1 Pgp MYLOTARG

Anti-CD33 3 Acve enedyne Plasma membrane antibody Calicheamicin derivave

CD33 Linker N-acetyl 4 gamma Nucleus

calicheamicin pH~6 2

pH~4 Cytoplasm

1. MYLOTARG binds to CD33 angens on leukaemic blasts 2. Once bound, the MYLOTARG/CD33 complex is internalised by receptor-mediated endocytosis 3. Calicheamicin is released from the anbody–drug complex and acts as a potent cytotoxic agent 4. Calicheamicin causes double-strand DNA breaks, causing the cell to undergo apoptosis ADC, anbody-drug conjugate; Pgp, P-glycoprotein Ricart AD. Clin Cancer 2011;17:6417–6427 12 Sievers E.L. et al JCO 2001 Sievers E.L. et al JCO 2001 1442

Final Report of the Efficacy and Safety of

1442 Gemtuzumab Ozogamicin (Mylotarg) in Patients with CD33-Positive Acute Myeloid Leukemia in First

Recurrence 1442 Final Report of the Efficacy and Safety of Gemtuzumab Ozogamicin (Mylotarg) in Patients with 1 Richard A. Larson, M.D. BACKGROUND. In this study, the authors analyzed the efficacy and safety of gem- 2,3 CD33-PositiveEric L. Sievers, M.D. AcuteFinaltuzumab Myeloid Report ozogamicin (GO) (Mylotarg of the௡), Leukemia an antibody-targeted Efficacy chemotherapy and for in Safety First of 4 Edward A. Stadtmauer, M.D. CD33-positive acute myeloid leukemia (AML). 5 Gemtuzumab Ozogamicin (Mylotarg) in Patients with RecurrenceBob Lo¨wenberg, M.D. METHODS. Patients with CD33-positive AML in first recurrence were entered in 3 6 Elihu H. Estey, M.D. open-label, single-arm, Phase II studies. Patients received monotherapy with GO 9 7 CD33-Positive2 Acute Myeloid Leukemia in First Herve´ Dombret, M.D. mg/m as a 2-hour intravenous infusion in 2 doses separated by 2 weeks. Patients 8 Matthias Theobald, M.D. Recurrencewere evaluated for remission, survival, and treatment-emergent adverse events. 9 RESULTS. Two hundred seventy-seven patients (median age, 61 yrs) were treated Dimitris Voliotis, M.D.1 Richard A. Larson, M.D. 10 with GO,BACKGROUND. and 71 patients (26%)In achieved this remission, study, the which authors was defined analyzed as Յ 5% the efficacy and safety of gem- John M. Bennett,2,3 M.D. Eric L. Sievers, M.D. 11 blasts intuzumab the bone marrow ozogamicin without leukemic (GO) blasts (Mylotargin the peripheral௡), blood, an neu- antibody-targeted chemotherapy for Maria Richie, B.S. 4 Edward A. Stadtmauer, M.D.11 trophil recoveryCD33-positive to Ն 1500/␮L, hemoglobin acute myeloidՆ 9 g/dL, and leukemia independence (AML). from red Lance H. Leopold,5 M.D. 1 M.D. Richard A. Larson,METHODS.M.D. Patients with CD33-positiveBACKGROUND. In AMLthis study, in the first authors recurrence analyzed the were efficacy entered and safety in of 3 gem- Bob Lo¨wenberg, 11 blood cell and platelet transfusions. Complete remission (CR) with platelet recov- Mark S. Berger,6 M.D. 2,3 Elihu H. Estey, M.D. Eric L. Sievers,open-label,M.D. single-arm, Phasetuzumab II studies. ozogamicin Patients (GO) (Mylotarg received௡), an monotherapy antibody-targeted with chemotherapy GO 9 for 11 ery (Ն 100,000/␮L) or without4 full platelet recovery (Ͻ 100,000/␮L) (CRp) was Matthew L. Sherman,7 M.D. Edward A. Stadtmauer,2 M.D. CD33-positive acute myeloid leukemia (AML). Herve´ Dombret, M.D. mg/m as a 2-hour intravenous infusion in 2 doses separated by 2 weeks. Patients 12 observed in 35 patients5 (13%) and 36 patients (13%), respectively. The median 8Ph.D. Bob Lo¨wenberg, M.D. METHODS. Patients with CD33-positive AML in first recurrence were entered in 3 Matthias Theobald,Michael R. Loken,M.D. were evaluated for remission, survival, and treatment-emergent adverse events. 13 recurrence-free survival6 was 6.4 months for patients who achieved CR and 4.5 9 Elihu H. Estey, M.D. open-label, single-arm, Phase II studies. Patients received monotherapy with GO 9 Dimitris Voliotis,Jacques J.M.D. M. van Dongen, M.D., Ph.D. RESULTS. Two hundred seventy-seven patients (median age, 61 yrs) were treated 2,3 months for patients7 who achieved CRp. Although expectedmg/m2 as incidences a 2-hour of intravenous Grade 3 infusion in 2 doses separated by 2 weeks. Patients 10 M.D. Herve´ Dombret,withM.D. GO, and 71 patients (26%) achieved remission, which was defined as Յ 5% John M. Bennett,Irwin D. Bernstein,M.D. 8 Matthiasor 4 neutropenia Theobald, (98%)M.D. and thrombocytopenia (99%)were were evaluated observed, for the remission, incidence survival, and treatment-emergent adverse events. 11 2 blasts in the bone marrow without leukemic blasts in the peripheral blood, neu- Maria Richie,FrederickB.S. R. Appelbaum, M.D. 9 Dimitrisof Grade Voliotis, 3 or 4M.D. sepsis (17%) and pneumonia (8%)RESULTS. was relativelyTwo low. hundred Grade 3 seventy-seven or 4 patients (median age, 61 yrs) were treated 11 Ն ␮ Ն for the MylotargM.D. Study Group trophil recovery10 to 1500/ L, hemoglobin 9 g/dL, and independence from red Lance H. Leopold, Johnhyperbilirubinemia M. Bennett, M.D. and hepatic aspartate aminotransferasewith GO, and and 71alanine patients amino- (26%) achieved remission, which was defined as Յ 5% 11 blood11 cell and platelet transfusions.blasts in the boneComplete marrow without remission leukemic (CR) blasts with in the platelet peripheral recov- blood, neu- Mark S. Berger, M.D. Mariatransferase Richie, B.S. elevations were reported in 29%, 18%, and 9% of patients, respectively; 1 Department of Medicine, University11 of Chicago, ery (Ն 100,000/11 ␮L) or withouttrophil recovery full platelet to Ն 1500/ recovery␮L, hemoglobin (Ͻ Ն100,000/9 g/dL, and␮L) independence (CRp) was from red Matthew L. Sherman, M.D. Lance0.9% H. of Leopold, patients whoM.D. did not undergo prior or subsequent hematopoietic stem cell Chicago, Illinois. 12 observed11 in 35 patients (13%)blood cell and and 36 platelet patients transfusions. (13%), Complete respectively. remission (CR) The with median platelet recov- Michael R. Loken, Ph.D. Marktransplantation S. Berger, M.D. developed hepatic venoocclusive disease after GO treatment. 13 recurrence-free11 survival wasery 6.4(Ն 100,000/ months␮L) or for without patients full platelet who recovery achieved (Ͻ 100,000/ CR and␮L) (CRp) 4.5 was Jacques J.2 M.Clinical van Research Dongen, Division, Fred HutchinsonM.D., Ph.D. Can- Matthew L. Sherman, M.D. 2,3 months for12 patients who achievedobserved in CRp. 35 patients Although (13%) and expected 36 patients incidences (13%), respectively. of Grade The median 3 Irwin D. Bernstein,cer Research Center,M.D. Seattle, Washington. Michael R. Loken, Ph.D. or 4 neutropenia (98%)13 andrecurrence-free thrombocytopenia survival was (99%) 6.4 months were for observed, patients who the achieved incidence CR and 4.5 2 M.D., Ph.D. Frederick R.3 Appelbaum, M.D. Jacques J. M. van Dongen, Department of Pediatrics, University of Washing- of Grade2,3 3 or 4 sepsis (17%)months and forpneumonia patients who achieved(8%) was CRp. relatively Although expected low. Grade incidences 3 ofor Grade 4 3 for the Mylotarg Study Group Irwin D. Bernstein, M.D. ton, Seattle, Washington. 2 or 4 neutropenia (98%) and thrombocytopenia (99%) were observed, the incidence Frederick R.hyperbilirubinemia Appelbaum, M.D. and hepatic aspartate aminotransferase and alanine amino- of Grade 3 or 4 sepsis (17%) and pneumonia (8%) was relatively low. Grade 3 or 4 4 Hematologic Malignancies Program, University of for the Mylotargtransferase Study Group elevations were reported in 29%, 18%, and 9% of patients, respectively; 1 Department of Medicine, University of Chicago, hyperbilirubinemia and hepatic aspartate aminotransferase and alanine amino- Pennsylvania Cancer Center, Philadelphia, Penn- 10 0.9% of patients who did not undergo prior or subsequent hematopoietic stem cell Chicago, Illinois. University of Rochester Medical Center, Roch- biostatisticaltransferase analysis, andSusan elevations Leinbach were for as- reported in 29%, 18%, and 9% of patients, respectively; sylvania. 1ester,Department New York. oftransplantation Medicine, University of Chicago, developed hepatic venoocclusive disease after GO treatment. sistance in article0.9% preparation. of patients They who also did thank not all undergo prior or subsequent hematopoietic stem cell 2 Clinical Research Division, Fred Hutchinson Can- Chicago, Illinois. 5 11 of the patients and clinical personnel who partic- cer Research Center,Department Seattle, of Hematology, Washington. Erasmus University Wyeth Research, Collegeville, Pennsylvania. transplantation developed hepatic venoocclusive disease after GO treatment. 2 ipated in these studies. Medical Center, Rotterdam, The Netherlands. Clinical Research Division, Fred Hutchinson Can- 12 3 Department of Pediatrics, TOSSICITA’ EPATICA G3-G4 DA MYLOTARG University of Washing- cerHematologics Research Center, Inc., Seattle, Seattle, Washington. Washington. 6 Additional investigators in the Mylotarg Study ton, Seattle, Washington.Department of Leukemia, The University of Texas 3 13DepartmentDepartment of Pediatrics,Immunology, University Erasmus University of Washing- M. D. Anderson Cancer Center, Houston, Texas. Group are M. Boogaerts, Universitair Ziekenhuis 4 ton, Seattle, Washington. Hematologic MalignanciesIPERBILIRUBINEMIA 29% Program, University of Medical Center, Rotterdam, The Netherlands. Gasthuisberg, Leuven Belgium; S. Castaigne, Hoˆ- 7 Hospital Saint Louis, Paris, France. 4 10 Pennsylvania Cancer Center, Philadelphia, Penn- SupportedHematologic inUniversity part Malignancies by research of Program,funding Rochester from University Wyeth Medical of pital A. Center, Mignot, LeRoch- Chesday, France;biostatistical P. Huijgens, analysis, and Susan Leinbach for as- sylvania. AUMENTO AST 18% Pennsylvania Cancer Center, Philadelphia, Penn- 10 8 Department of Hematology, Johannes Gutenberg Pharmaceuticalsester, toNew all of York.the clinical investigators. AcademischUniversity Ziekenhuis of Rochester VU, Amsterdam, Medicalsistance The Neth-Center, in Roch- article preparation.biostatistical analysis, They and also Susan thank Leinbach all for as- 5 sylvania. erlands;ester, R. Spielberger, New York. City of Hope Nationalof the Med- patients andsistance clinical in article personnel preparation. who They partic- also thank all Department ofUniversity, Hematology, Mainz, Germany. Erasmus University 11 Wyeth Research, Collegeville, Pennsylvania. 5 ipated in these studies.of the patients and clinical personnel who partic- Medical Center, Rotterdam,AUMENTO ALT 9% The Netherlands. TheDepartment authors thank ofHematology, Vincent H. J. van Erasmus der Velden University for ical Center,11 Wyeth Duarte, Research, CA; M.Collegeville, Tallman, Northwestern Pennsylvania. 9 12 ipated in these studies. Department of Oncology, University Hospital Co- MedicalCD33/Mylotarg Center,Hematologics laboratory Rotterdam, analysis, The Inc., Netherlands. Nadine Seattle, Dodge Washington.University Feinberg School of Medicine, Chicago, 6 12 Hematologics Inc., Seattle,Additional Washington. investigators in the Mylotarg Study Department oflogne, Leukemia, Cologne, Germany. The University of Texas for clinical13 programming, Robert Herbertson for IL; C. Bernasconi. Istituto di Ematolgia, Policlinico VOD/SOS 5%, MA 17% NEI PAZIENTI 6 DepartmentDepartment of Leukemia, The of University Immunology, of Texas Erasmus University Additional investigators in the Mylotarg Study M. D. Anderson Cancer Center, Houston, Texas. 13 Department of Immunology,Group Erasmus are University M. Boogaerts, Universitair Ziekenhuis M. D. AndersonMedical Cancer Center, Center, Rotterdam, Houston, Texas. The Netherlands. Group are M. Boogaerts, Universitair Ziekenhuis Medical Center, Rotterdam, TheGasthuisberg, Netherlands. Leuven Belgium; S. Castaigne, Hoˆ- 7 Hospital Saint© Louis, 2005 American Paris, Cancer France. Society Gasthuisberg, Leuven Belgium; S. Castaigne, Hoˆ- SUCCESSIVAMENTE TRATTATI CON ALLOBMT 7 HospitalSupported Saint Louis, in Paris, part France. by research funding from Wyeth pital A. Mignot, Le Chesday, France; P. Huijgens, DOI 10.1002/cncr.21326 Supported in part by research funding from Wyeth pital A. Mignot, Le Chesday, France; P. Huijgens, 8 Pharmaceuticals to all of the clinical investigators. Academisch Ziekenhuis VU, Amsterdam, The Neth- Department ofPublished Hematology, online 22 August Johannes 2005 in Wiley Gutenberg InterScience8 (www.interscience.wiley.com). Pharmaceuticals to all of the clinicalLarson R.A. et al Cancer 2005 investigators. Academisch Ziekenhuis VU, Amsterdam, The Neth- Department of Hematology, Johannes Gutenberg erlands; R. Spielberger, City of Hope National Med- University, Mainz, Germany. University, Mainz, Germany. erlands; R. Spielberger, City of Hope National Med- The authors thank Vincent H. J. van der Velden for ical Center, Duarte, CA; M. Tallman, Northwestern 9 The authors thank Vincent H. J. van der Velden for ical Center, Duarte, CA; M. Tallman, Northwestern Department of Oncology, University Hospital Co- 9 DepartmentCD33/Mylotarg of Oncology, University laboratory Hospital analysis, Co- CD33/Mylotarg Nadine Dodge laboratory analysis,University Nadine Dodge FeinbergUniversity School Feinberg of Medicine, School of Medicine, Chicago, Chicago, logne, Cologne, Germany. logne, Cologne,for clinical Germany. programming, Robertfor Herbertson clinical programming, for RobertIL; C. Herbertson Bernasconi. for IstitutoIL; C. Bernasconi. di Ematolgia, Istituto di Ematolgia, Policlinico Policlinico

© 2005 American Cancer Society © 2005 American Cancer Society DOI 10.1002/cncr.21326 DOI 10.1002/cncr.21326 Published online 22 August 2005 in Wiley InterSciencePublished (www.interscience.wiley.com). online 22 August 2005 in Wiley InterScience (www.interscience.wiley.com). GO-ASSOCIATEDLetter to the Editor / Leukemia Research SINUSOIDAL35 (2011) e84–e86 e85 OBSTRUCTIVE SYNDROME (SOS) POSSIBILI MECCANISMI PATOGENETICI • DANNO SULLE CELLULE ENDOTELIALI SINUSOIDALI EPATICHE DA PARTE DELLA CALICOMICINA UNA VOLTA STACCATASI DALL’ANTICORPO • UPTAKE NON-SPECIFICO DEI COMPLESSI ANTICORPO-CALICOMICINA DA PARTE DELLE CELLULE DEL KUPFFER • DEPLEZIONE DEL GLUTATIONE NELLE CELLULE ENDOTELIALI SINUSOIDALI

Fig. 1. Biochemical signs of hepatotoxicity after GO treatment and CD33 expression on Kupffer cells and hepatocytes. (A) Time course of ALAT concentrations showing a 5-fold increase less than two weeks after administration of the second GO dose. “GO” represents time of GO treatment and red arrowheads represent normal ALAT values (5-45 U/l). (B) Double immunohistochemical staining performed on stored formalin-fixed, paraffin-embedded liver tissue from a healthy individual. Double staining with CD33 (fast red precipitates; violet colored) and CD163 (DAB precipitates; brown colored) showing CD163 immunoreactivity limited to the constituent Kupffer cells (marked with grey arrowheads), which also express the myeloid lineage marker CD33 (marked with white arrowheads). Importantly, the staining shows that CD33 is highly expressed on hepatocytes (marked with black arrowheads) (Mayer’s hematoxylin counterstain, original magnification 100, oil). The figure is representative of several stainings of × liver tissue from different healthy individuals. (C) An isotype-matched negative control antibody was used to evaluate non-specific binding of anti-CD33.

Conflict of interest [7] LowenbergManiecki M.B. et al Leuk Research 2011 B, Beck J, Graux C, van PW, Schouten HC, Verdonck LF, et al. Gemtuzumab ozogamicin as postremission treatment in AML at 60 years of age or more: results of a multicenter phase 3 study. Blood 2010;115(13): The authors declare no conflict of interest. 2586–91. [8] McKoy JM, Angelotta C, Bennett CL, Tallman MS, Wadleigh M, Evens AM, et al. Acknowledgements Gemtuzumab ozogamicin-associated sinusoidal obstructive syndrome (SOS): an overview from the research on adverse drug events and reports (RADAR) project. Leuk Res 2007;31(5):599–604. The authors wish to thank Tom Nordfeld Troelsen, Institute of [9] Larson RA, Sievers EL, Stadtmauer EA, Lowenberg B, Estey EH, Dombret H, et al. Pathology, Aarhus University Hospital, for excellent technical assis- Final report of the efficacy and safety of gemtuzumab ozogamicin (Mylotarg) in patients with CD33-positive acute myeloid leukemia in first recurrence. Cancer tance. This work was supported by grants from the Danish Medical 2005;104(7):1442–52. Research Council, The Novo Nordisk Foundation, and The Danish [10] Maniecki MB, Hasle H, Friis-Hansen L, Lausen B, Nielsen OJ, Bendix Cancer Society (R2-A198-09-S2). K, et al. Impaired CD163-mediated hemoglobin-scavenging and severe toxic symptoms in patients treated with gemtuzumab ozogamicin. Blood 2008;112(4):1510–4. References Maciej Bogdan Maniecki ∗ [1] Voutsadakis IA. Gemtuzumab Ozogamicin (CMA-676, Mylotarg) for the treat- Department of Clinical Biochemistry, Aarhus ment of CD33+ acute myeloid leukemia. Anticancer Drugs 2002;13(7):685–92. [2] Sievers EL. Efficacy and safety of gemtuzumab ozogamicin in patients with Sygehus, Aarhus University Hospital, Nørrebrogade CD33-positive acute myeloid leukaemia in first relapse. Expert Opin Biol Ther 44, DK-8000 Aarhus C, Denmark 2001;1(5):893–901. [3] Sievers EL, Linenberger M. Mylotarg: antibody-targeted chemotherapy comes Henrik Hasle of age. Curr Opin Oncol 2001;13(6):522–7. Department of Pediatrics, Aarhus University Hospital, [4] Bross PF, Beitz J, Chen G, Chen XH, Duffy E, Kieffer L, et al. Approval summary: gemtuzumab ozogamicin in relapsed acute myeloid leukemia. Clin Cancer Res Skejby, DK-8200, Aarhus N, Denmark 2001;7(6):1490–6. [5] Arceci RJ, Sande J, Lange B, Shannon K, Franklin J, Hutchinson R, et al. Safety Knud Bendix and efficacy of gemtuzumab ozogamicin in pediatric patients with advanced University Institute of Pathology, Aarhus Sygehus, CD33+ acute myeloid leukemia. Blood 2005;106(4):1183–8. Aarhus University Hospital, Tage-Hansen Gade 2, [6] Burnett AK, Hills RK, Milligan D, Kjeldsen L, Kell J, Russell NH, et al. Identification DK-8000, Aarhus C, Denmark of patients with acute myeloblastic leukemia who benefit from the addition of gemtuzumab ozogamicin: results of the MRC AML15 trial. J Clin Oncol 2010. MAGGIO 2000

• FDA-ACCELERATED APPROVAL OF MYLOTARG FOR THE TREATMENT OF PATIENTS WITH CD33+ AML IN FIRST RELAPSE WHO ARE ≥ 60-YRS OLD AND WHO ARE NOT CONSIDERED CANDIDATES FOR CYTOTOXIC CHEMOTHERAPY

• THE FIRST ANTIBODY-DRUG CONJUGATE APPROVED FOR CANCER THERAPY AND THE FIRST TARGETED AGENT IN NON-M3 AML SUBGROUP From www.bloodjournal.org by guest on May 8, 2018. For personal use only. Regular Article

CLINICAL TRIALSFrom AND www.bloodjournal.org OBSERVATIONS by guest on May 8, 2018. For personal use only. Regular Article Aphase3studyofgemtuzumabozogamicinduringinductionand postconsolidationCLINICAL TRIALS ANDtherapy OBSERVATIONS in younger patients with acute myeloidAphase3studyofgemtuzumabozogamicinduringinductionand leukemia

postconsolidation1 therapy1,2 in younger3 patients with4 acute 5 6 Stephenmyeloid H. Petersdorf, leukemiaKenneth J. Kopecky, Marilyn Slovak, Cheryl Willman, Thomas Nevill, Joseph Brandwein, Richard A. Larson,7 Harry P. Erba,8 Patrick J. Stiff,9 Robert K. Stuart,10 Roland B. Walter,1 Martin S. Tallman,11 Leif Stenke,12 1 1,2 3 4 5 6 Stephen H. Petersdorf, Kenneth1 J. Kopecky, Marilyn Slovak, Cheryl Willman, Thomas Nevill, Joseph Brandwein, and FrederickRichard A. R. Larson, Appelbaum7 Harry P. Erba,8 Patrick J. Stiff,9 Robert K. Stuart,10 Roland B. Walter,1 Martin S. Tallman,11 Leif Stenke,12 and Frederick R. Appelbaum1 1Fred Hutchinson Cancer Research Center, Seattle, WA; 2Southwest Oncology Group Statistical Center, Seattle, WA; 3City of Hope Medical Center, Duarte, CA; 4 1Fred Hutchinson Cancer Research Center,5 Seattle, WA; 2Southwest Oncology Group Statistical Center, Seattle,6 WA; 3City of Hope Medical Center, Duarte, CA; University4 of New Mexico, Albuquerque, NM; Vancouver5 General Hospital, Vancouver, BC, Canada;6 University of Toronto, Toronto, Ontario, Canada; 7 University of New Mexico, Albuquerque,8 NM; Vancouver General Hospital,9 Vancouver, BC, Canada; University of Toronto, Toronto,10 Ontario, Canada; University7University of Chicago, of Chicago, Chicago, Chicago, IL; University IL; 8University of Michigan, of Michigan, Ann Ann Arbor, Arbor, MI; MI; Loyola9Loyola University University Medical Medical Center, Center, Maywood, Maywood, IL; IL;10MedicalMedical University University of South of South Carolina,Carolina, Charleston, Charleston, SC; 11 SC;Memorial11Memorial Sloan-Kettering Sloan-Kettering Cancer Cancer Center, Center, New New York, York, NY; NY; and and 12Karolinska University University Hospital, Hospital, Stockholm, Stockholm, Sweden Sweden

Key Points This randomized phase 3 clinical trial evaluated the potential benefit of the addition of Thisgemtuzumab randomized ozogamicin phase 3 clinical (GO) to trialstandard evaluated induction the and potential postconsolidation benefit of the therapy addition in of • TheKey addition Points of gemtuzumab patients with acute myeloid leukemia. Patients were randomly assigned to receive gemtuzumab ozogamicin2 (GO) to standard induction and postconsolidation2 therapy in ozogamicin to induction or daunorubicin (45 mg/m per day on days 1, 2, and 3), cytarabine (100 mg/m per day by • The additionmaintenance of gemtuzumab therapy failed patientscontinuous with infusion acute myeloid on days leukemia. 1–7), and GO Patients (6 mg/m were2 on randomlyday 4; DA1 assignedGO) vs standard to receive 2 ozogamicinto improve to induction the complete or daunorubicininduction therapy (45 mg/m with2 daunorubicinper day on (60 days mg/m 1, 2,per and day 3), on cytarabine days 1, 2, and (100 3) and mg/m cytarabine2 per day by alone (DA). Patients who achieved complete remission (CR) received 3 courses of high- response rate or overall continuous infusion on days 1–7), and GO (6 mg/m2 on day 4; DA1GO) vs standard maintenancesurvival in therapy patients failed with acute dose cytarabine. Those remaining in CR after consolidation were randomly assigned to inductionreceive therapy either no with additional daunorubicin therapy (60 or 3mg/m doses2 per of dayGO (5on mg/m days2 1,every 2, and 28 3) days). and cytarabine From to improvemyeloid the leukemia. complete aloneAugust (DA). 2004 Patients until August who achieved 2009, 637 complete patients were remission registered (CR) for received induction. 3 courses The CR rate of high- responsewas 69% for rate DA1 orGO overall and 70% for DA (P 5 .59). Among those who achieved a CR, the 5-year relapse-free survival rate was 43% in the survivalDA1GO groupin patients and 42% with in the acute DA groupdose (P 5 .40). cytarabine. The 5-year Those overall remaining survival rate in CR was after 46% consolidation in the DA1GO groupwere randomly and 50% in assigned the DA to group (P 5 .85). One hundred seventy-fourreceive patients either in no CR additional after consolidation therapy or underwent 3 doses the of GOpostconsolidation (5 mg/m2 every randomization. 28 days). From myeloidDisease-free leukemia. survival was not improved with postconsolidation GO (HR, 1.48; P 5 .97). In this study, the addition of GO to induction or postconsolidation therapy failed to showAugust improvement 2004 in until CR rate, August disease-free 2009, 637 survival, patients or overall were registered survival. This for trial induction. is registered The with CR rate was 69%www.clinicaltrials.gov for DA1GO and 70% as #NCT00085709. for DA (P 5 .59). (Blood Among. 2013;121(24):4854-4860) those who achieved a CR, the 5-year relapse-free survival rate was 43% in the DA1GOIntroduction group and 42% in the DA group (P 5 .40). The 5-year overall survival rate was 46% in the DA1GO group and 50% in the DA group (P 5 .85). One hundred seventy-four patients in CR after consolidation underwent the postconsolidation randomization. Standard induction therapy for acute myeloid leukemia (AML) is Initial phase 2 data for this agent showed promise for patients treated Disease-freea combination survival of was cytarabine not improved and an anthracycline. with postconsolidation For the last GO 30 (HR,in fi 1.48;rst relapse.P 5 .97). Among In this 142 study, CD33-positive the addition patients of GO with to recurrent induction or postconsolidationyears, there has therapy been only failed limited to show improvement improvement in complete in CR remis- rate, disease-freeAML treated survival, with 2 doses or overall of GO, survival. 23 patients This achieved trial is registered CR and 19 with www.clinicaltrials.govsion (CR) rates and as overall #NCT00085709. survival (OS) (withBlood chemotherapy,. 2013;121(24):4854-4860) and the achieved CR with incomplete platelet recovery, for an overall re- improvements that have occurred are primarily the result of dose sponse rate of 30%.10,11 These results led to the accelerated approval escalation of standard agents during induction and consolidation and of the drug by the US Food and Drug Administration (FDA) for Introductionimprovements in supportive care.1-4 For patients younger than 60 treatment of patients older than 60 years with AML in first relapse years, a CR is typically obtained in 65% to 80% of patients, but the who were not candidates for aggressive chemotherapy. Standardmajority induction of these therapy patients for will acute relapse myeloid if treated leukemia with standard (AML) con- is InitialThe phase availability 2 data for of thisGO agent prompted showed further promise investigation for patients of this treated solidation chemotherapy. agentfi in combination with chemotherapy. Although the approved a combinationThe majority of cytarabine of AML and cells an express anthracycline. the CD33 For surface the last antigen, 30 indoserst of relapse. GO was Among 9 mg/m2 142given CD33-positive twice 14 days apart,patients initial with studies recurrent years, therewhich has is beennot expressed only limited on normal improvement hematopoietic in complete stem remis-cells or AMLdemonstrated treated withconsistent 2 doses saturation of GO, of 23 CD33 patients receptors achieved at a dose CR andof 19 sion (CR)nonhematopoietic rates and overall cells. survival5,6 Initial (OS) trials with of chemotherapy, radiolabeled anti-CD33 and the achieved6 mg/m2.9 CRA phase with 1/2 incomplete trial, W-R platelet 206, was recovery, undertaken for to deanfi overallne the re- antibodies showed that the antigen rapidly internalized after antibody maximum tolerated10,11 dose ofPetersdorf S.H. et al Blood 2013 daunorubicin and cytarabine (DA) ad- improvementsbinding.7,8 thatThese have occurred observations are primarily suggested the that result an antibody of dose– sponseministered rate with of 30%. a dose ofThese GO known results to led saturate to the accelerated CD33 receptors approval escalationchemotherapy of standard immunoconjugate agents during induction targeted and to CD33consolidation might be and an of(6 the mg/m drug2). The by maximally the US Food tolerated and doses Drug were Administration estimated to be (FDA) dau- for 2 improvementseffective in way supportive to treat AML. care.1-4 GemtuzumabFor patients ozogamicin younger (GO) than 60was treatmentnorubicin 45of patientsmg/m per older day than on days 60 years 1 through with 3 AML and cytarabine in first relapse developed, consisting of a humanized anti-CD33 monoclonal antibody 100 mg/m2 on days 1 through 7, with GO 6 mg/m2 on day 4. years, aconjugated CR is typically to calicheamicin, obtained in a65% highly to potent 80% ofantitumor patients, antibiotic. but the 9 whoAmulti-institutionalphase2trialwasopenedinOctober2001, were not candidates for aggressive chemotherapy. majority of these patients will relapse if treated with standard con- The availability of GO prompted further investigation of this Submitted January 10, 2013; accepted February 22, 2013. Prepublished The publication costs of this article were defrayed in part by page charge solidationonline chemotherapy. as Blood First Edition paper, April 16, 2013; DOI 10.1182/blood-2013- agentpayment. in Therefore,combination and solely with to chemotherapy. indicate this fact, Although this article the is hereby approved The01-466706. majority of AML cells express the CD33 surface antigen, dosemarked of “advertisement” GO was 9 mg/m in accordance2 given with twice 18 USC 14 days section apart, 1734. initial studies whichThere is not is an expressed Inside Blood oncommentary normal on hematopoietic this article in this stem issue. cells or demonstrated© 2013 by The American consistent Society saturation of Hematology of CD33 receptors at a dose of nonhematopoietic cells.5,6 Initial trials of radiolabeled anti-CD33 6 mg/m2.9 A phase 1/2 trial, W-R 206, was undertaken to define the antibodies4854 showed that the antigen rapidly internalized after antibody maximum toleratedBLOOD, dose of 13 daunorubicin JUNE 2013 x VOLUME and cytarabine 121, NUMBER (DA) 24 ad- binding.7,8 These observations suggested that an antibody– ministered with a dose of GO known to saturate CD33 receptors chemotherapy immunoconjugate targeted to CD33 might be an (6 mg/m2). The maximally tolerated doses were estimated to be dau- effective way to treat AML. Gemtuzumab ozogamicin (GO) was norubicin 45 mg/m2 per day on days 1 through 3 and cytarabine developed, consisting of a humanized anti-CD33 monoclonal antibody 100 mg/m2 on days 1 through 7, with GO 6 mg/m2 on day 4. conjugated to calicheamicin, a highly potent antitumor antibiotic.9 Amulti-institutionalphase2trialwasopenedinOctober2001,

Submitted January 10, 2013; accepted February 22, 2013. Prepublished The publication costs of this article were defrayed in part by page charge online as Blood First Edition paper, April 16, 2013; DOI 10.1182/blood-2013- payment. Therefore, and solely to indicate this fact, this article is hereby 01-466706. marked “advertisement” in accordance with 18 USC section 1734.

There is an Inside Blood commentary on this article in this issue. © 2013 by The American Society of Hematology

4854 BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 From www.bloodjournal.org by guest on May 8, 2018. For personal use only.

BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 GEMTUZUMAB OZOGAMICIN IN AML INDUCTION/MAINTENANCE 4855

Table 1. Pretreatment characteristics of 595 adult patients with or possibly both. Study S0106 was an intergroup study with patient previously untreated AML, by treatment group enrollment from several cooperative groups including SWOG, the DA1GO (n 5 295) DA (n 5 300) National Cancer Institute of Canada, the Leukemia Group of Middle Median Min–Max Median Min–Max P* Sweden/the Swedish AML Group, Cancer and Acute Leukemia Age, years 47 18-60 48 18-60 .44 Group B, and the Eastern Cooperative Oncology Group. White blood cells, 10.7 0.5-545.0 12.5 0.2-243.5 .48 109/L Peripheral blood blasts, % 34 0-99 27 0-99 .16 (n 5 555) Methods Neutrophils, % 9 0-97 10 0-72 .66 (n 5 574) Patient population Absolute neutrophil 1.1 0-171.6 0.9 0-40.1 .32 Patients with AML according to the World Health Organization criterion 9 count, 10 /L (>20% blasts), aged 18 to 60 years, and with a Zubrod performance score (n 5 574) of from 0 to 3 and adequate organ function (bilirubin <2 3 institutional Hemoglobin, g/dL 9.1 3.5-18.0 9.1 4.4-29.1 .81 upper limit of normal, serum glutamic oxaloacetic transaminase and serum (n 5 583) glutamic pyruvate transaminase <3 3 institutional upper limit of normal, 9 Platelets, 10 /L 53 2-7900 55 7-9300 .39 and left ventricular ejection fraction > 50%) were eligible. Patients with (n 5 593) acute promyelocytic leukemia (M3 AML), unstable cardiac arrhythmias or Bone marrow blasts, % 66 7-100 65 3-100 .72 angina, or known hepatitis B or active hepatitis C were not eligible. Prior in (n 5 584) situ cervical carcinoma or adequately treated prior basal or squamous cell Patients % Patients % skin cancer or stage I or II cancer in remission were permitted, as was any Age, years prior cancer from which the patient was disease-free for 5 years. Patients ,35 57 19% 56 19% .92 with AML arising from a prior hematological malignancy were ineligible. $35 238 81% 244 81% One prior dose of intrathecal chemotherapy for acute leukemia was permitted, Sex From www.bloodjournal.org by guest on May 8, but2018. patients For couldpersonal not have use received only. prior systemic chemotherapy for Female 135 46% 147 49% .46 leukemia. All patients provided written informed consent in accordance BLOOD, 13Male JUNE 2013 x VOLUME 121, NUMBER 160 24 54% 153 51% GEMTUZUMABwith local OZOGAMICIN policies, federal IN regulations, AML INDUCTION/MAINTENANCE and the Declaration of Helsinki. 4855 French-American British classification Study design and treatment groups Table 1. PretreatmentM1 characteristics 67of 595 adult 23% patients 58 with 20% .76or possibly both. Study S0106 was an intergroup study with patient Patients were initially randomly assigned 1:1 between 2 induction regimens: previouslyM2 untreated AML, by treatment 76 group 26% 68 24% enrollment from several cooperative groups including SWOG, the either DA1GO, daunorubicin 45 mg/m2 by IV push on days 1 through 3, M4DA1GO (n 735 295) 25% DA (n 715 300) 25% National Cancer Institute of Canada, the Leukemia Group of Middle cytarabine 100 mg/m2 by continuous IV infusions on days 1 through 7, and M4eosMedian Min–Max 9 3% Median 10 Min–Max 3%P* Sweden/the Swedish AML Group,2 Cancer and Acute Leukemia gemtuzumab ozogamicin 6 mg/m by 2-hour IV infusion on day 4; or DA, M5 38 13% 47 16% Group B, and the Eastern2 Cooperative Oncology Group. Age, years 47 18-60 48 18-60 .44 daunorubicin 60 mg/m by IV push on days 1 through 3 and cytarabine White blood cells,M610.7 0.5-545.0 4 1% 12.5 0.2-243.5 9 3% .48 109/L M7 3 1% 3 1% Table 2. Pretreatment cytogenetic characteristics of 496 adult Peripheral bloodM0 blasts, % 34 21 0-99 7% 27 23 0-99 8% .16 patients with previously untreated AML, by treatment group (n 5 555) Unknown 4 — 11 — Methods DA1GO (n 5 254) DA (n 5 242) Neutrophils,Performance % status 9 0-97 10 0-72 .66 Patient population Patients % Patients % P* (n 5 574) 011740%11840%.37 Absolute neutrophil 1.1 0-171.6 0.9 0-40.1 .32 114750%13646%PatientsRisk with group AML according to the World Health Organization criterion count, 109/L 2227%3110%(>20%Favorable blasts), aged 18 to 60 years,37 and with15 a Zubrod44 performance 18 score .47 (n 5 574) 383%134%of fromIntermediate 0 to 3 and adequate137 organ function54 (bilirubin 132<2 3 institutional 55 Hemoglobin, g/dL 9.1 3.5-18.0 9.1 4.4-29.1 .81 Unknown 1 — 2 — upper limitUnfavorable of normal, serum glutamic62 oxaloacetic24 transaminase55 23 and serum (n 5 583) glutamicIndeterminate pyruvate transaminase18<3 3 institutional7 upper11 limit of 5 normal, Platelets, 109/L 53 2-7900 55 7-9300 .39 *Two-sided P value from Wilcoxon test (continuous variables), Fisher’s exact testand leftNormal ventricular ejection fraction106 > 50%)45 were eligible. 103 Patients 46 with .85 (n 5 593)(age group, sex), or Pearson’s x-square test (French-American British classification,acute promyelocytic leukemia (M3 AML), unstable cardiac arrhythmias or CBF† 31 13 40 18 .20 Bone marrowperformance blasts, % status). 66 7-100 65 3-100 .72 angina, or known hepatitis B or active hepatitis C were not eligible. Prior in inv(16) 17 7 23 10 .32 (n 5 584) situ cervical carcinoma or adequately treated prior basal or squamous cell t(8;21) 14 6 17 8 .58 evaluating these doses.Patients Of 43 evaluable % patients, Patients 37 (84%) % achievedskin cancer or stage I or II cancer in remission were permitted, as was any 27, 7q, 25or25q 29 12 22 10 .46 Age, yearsCR. The incidence of elevated liver function tests including aspartateprior cancer from which the patient was disease-free for 5 years. Patients 27, 7q– 24 10 15 7 .19 ,35 57 19% 56 19% .92 with AML arising from a prior hematological malignancy were ineligible. aminotransferase (2%), alanine aminotransferase (2%), and bilirubin 25, 5q– 14 6 14 6 1.00 $3512 238 81% 244 81% One prior dose of intrathecal chemotherapy for acute leukemia was permitted, Sex (9%) was acceptable. but patients18 could not have received28 prior12 systemic19 chemotherapy 9 for .28 FemaleGiven the manageable 135 toxicity 46% of 147 this combination 49% .46 withleukemia.11q23 All patients provided11 written informed5 consent13 in accordance 6 .68 Malepromising efficacy in 160 the phase 54% 2 trial, the153 Southwest 51% Oncologywith2 local17 policies, federal regulations,9 and4 the Declaration63.60 of Helsinki. French-AmericanGroup British(SWOG) initiated study S0106 to compare in a prospective 218 6 3 521.00 Study design and treatment groups classificationrandomized trial the effects of adding GO to standard induc- t(9;11) 5 2 421.00 M1 67 23% 58 20% .76 Patientst(6;9) were initially randomly4 assigned 1:12 between 22 induction 1 regimens: .69 M2tion therapy with DA alone. 76 To ensure26% adequate 68 anthracycline 24% dose eitherinv(3) DA1GO, daunorubicin 453 mg/m2 by1 IV push on311.00 days 1 through 3, M4intensity in the control group, 73 this 25% protocol employed 71 25% daunorubicin at 2 2 2 cytarabine21q22 100 mg/m by continuous3 IV infusions1 on days311.00 1 through 7, and M4eos60 mg/m on days 1 through 9 3 with 3% cytarabine 10 at 100 3% mg/m per 2 gemtuzumabMarker/ring ozogamicin 6 mg/m17 by 2-hour7 IV infusion94.16 on day 4; or DA, M5 38 13% 47 16% 2 day by continuous infusion on days 1 through 7. In addition,daunorubicinComplex‡ 60 mg/m by IV36 push on days15 1 through34 3 and 15 cytarabine 1.00 M6the protocol included a4 second 1% randomization 9 to 3% test whether M7 3 1% 3 1% Other abnormality 79 34 74 33 .92 administration of GO after consolidation therapy would improveTable 2. Pretreatment cytogenetic characteristics of 496 adult M0 21 7% 23 8% patients with previously untreated AML, by treatment group *Two-sided P value from Pearson’s x-square test (Risk group) or Fisher’s exact Unknowndisease-free survival (DFS). 4 These — 2 randomizations 11 — were designed test (normal or specificDA abnormalities,1GO (n 5 based254) on 234 DA DA1GO (n and5 242) 223 DA patients. Performanceto determine status whether any beneficial effect from GO was achieved by †Core binding factor.Patients % Patients % P* 011740%11840%.37administration during either induction or postconsolidation therapy, ‡Three or more clonal cytogenetic abnormalities. 114750%13646%Risk group 2227%3110%Favorable 37 15 44 18 .47 383%134%Intermediate 137 54 132 55 Unknown 1 — 2 — Unfavorable 62 24 55 23 Indeterminate 18 7 11 5 *Two-sided P value from Wilcoxon test (continuous variables), Fisher’s exact test Normal 106 45 103 46 .85 (age group, sex), or Pearson’s x-square test (French-American British classification, CBF† 31 13 40 18 .20 performance status). inv(16) 17 7 23 10 .32 t(8;21) 14Petersdorf S.H. et al Blood 2013 6 17 8 .58 evaluating these doses. Of 43 evaluable patients, 37 (84%) achieved 27, 7q, 25or25q 29 12 22 10 .46 CR. The incidence of elevated liver function tests including aspartate 27, 7q– 24 10 15 7 .19 aminotransferase (2%), alanine aminotransferase (2%), and bilirubin 25, 5q– 14 6 14 6 1.00 12 (9%) was acceptable. 18 28 12 19 9 .28 Given the manageable toxicity of this combination with 11q23 11 5 13 6 .68 promising efficacy in the phase 2 trial, the Southwest Oncology 217 9 4 63.60 Group (SWOG) initiated study S0106 to compare in a prospective 218 6 3 521.00 randomized trial the effects of adding GO to standard induc- t(9;11) 5 2 421.00 tion therapy with DA alone. To ensure adequate anthracycline dose t(6;9) 4 2 2 1 .69 intensity in the control group, this protocol employed daunorubicin at inv(3) 3 1 311.00 21q22 3 1 311.00 60 mg/m2 on days 1 through 3 with cytarabine at 100 mg/m2 per Marker/ring 17 7 94.16 day by continuous infusion on days 1 through 7. In addition, Complex‡ 36 15 34 15 1.00 the protocol included a second randomization to test whether Other abnormality 79 34 74 33 .92 administration of GO after consolidation therapy would improve disease-free survival (DFS). These 2 randomizations were designed *Two-sided P value from Pearson’s x-square test (Risk group) or Fisher’s exact fi test (normal or specific abnormalities, based on 234 DA1GO and 223 DA patients. to determine whether any bene cial effect from GO was achieved by †Core binding factor. administration during either induction or postconsolidation therapy, ‡Three or more clonal cytogenetic abnormalities. From www.bloodjournal.org by guest on May 8, 2018. For personal use only.

4856 PETERSDORF et al BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24

Table 3. Treatment outcomes following induction chemotherapy of 595 adult patients with previously untreated AML, by treatment group CR CR or Cri Resistant disease OS at 5 years RFS at 5 years Group Patients %95%CI%95%CI%95%CI%95%CI%95%CI

DA1GO 295 69 63-74 76 69-79 15 12-20 46 40-52 43 36-50 DA 300 70 64-75 74 69-79 20 16-25 50 44-56 42 35-49 P* .59 .36 .065 .85 .40

*One-sided P value for superior outcome (higher CR rate, lower RD, HR,1) in DA1GO group, based on Fisher’s exact test (CR, RD) or logrank test (OS, RFS).

100 mg/m2 by continuous IV infusion on days 1 through 7. Marrow response CR/CRi/PR, and RD. Logrank tests and proportional hazards regression was assessed on day 14, and if there was aplasia with less than 5% blasts, were used to analyze OS, RFS, and DFS. All HR values comparing treat- growth factor therapy (sargramostim (rhGM-CSF), filgrastim (G-CSF) or ment groups are for DA1GO relative to DA or GO relative to observation; pegfilgrastim) could begin at the treating physician’s discretion. For both therefore, an HR lower than 1 indicates a superior outcome in the GO- groups, a second course using the DA regimen was allowed for patients with containing group. The statistical significance of treatment effects on marrow having more than 20% cellularity and more than 40% blasts on day response, OS, DFS, and RFS is represented by 1-sided P values for superior 14, or with 5% or more blasts on a subsequent examination. The induction outcomes in the DA1GO induction or GO maintenance groups; all other randomization was stratified by patient age (,35 years vs >35 years). P values are 2-sided. Confidence intervals (CIs) are at the 95% confidence Patients who achieved CR and were afebrile and free of infection, with level. The following results were based on data available February 3, 2013. adequate organ function, performance status, and resolution of any central nervous system involvement, were eligible to receive 3 courses of consolidation Interim analyses and early closure therapy with cytarabine 3 g/m2 by 3-hour continuous IV infusion every 12 This study was monitored by the SWOG Data and Safety Monitoring hours on days 1, 3, and 5. Consolidation courses were administered monthly. Committee (DSMC). Interim analyses of induction results were scheduled After completing consolidation therapy, patients who continued to meet when 228 and 456 patients were evaluated for response to induction the criteria for consolidation and who had not experienced sinusoidal ob- chemotherapy. Three interim analyses of postconsolidation DFS were structive syndrome during or after induction therapy were eligible for scheduled when 25%, 50%, and 75% of the expected number of events postconsolidation randomization (1:1) between GO (5 mg/m2, 3 doses at occurred, respectively. On August 11, 2009, the DSMC reviewed the least 28 days apart) vs observation. The postconsolidation randomization second scheduled interim analysis of CR rates, which was based as planned was stratified by prior use of GO (yes vs no) and preinduction cytogenetic on the first 456 evaluable patients. The CR rates in that analysis were 66% risk group (favorable vs intermediate vs unfavorable vs indeterminate). in 227 patients in the DA1GO group and 69% in 229 patients in the DA Patients were required to have an absolute neutrophil count higher than group, and the hypothesis that the DA1GO regimen increases the CR 1000/mm3 and a platelet count higher than 100 000/mm3 to receive each rate by 12% was rejected at the predefined significance level (P , .0025). cycle of postconsolidation GO. Additional analyses showed that RFS was not significantly better on the DA1 GO group. The DSMC also reviewed the first planned interim analysis of Treatment outcomes postconsolidation DFS. That analysis rejected the hypothesis that GO fi CR, CR with incomplete hematologic recovery (CRi), partial response (PR), improves DFS, with a hazard ratio (observation: GO) of 1.5 at the prespeci ed fi and resistant disease (RD) were defined according to the International Working signi cance level (P , .001). On the basis of these results, as well as the higher Group Guidelines.13 DFS was measured from the day of postconsolidation incidence of fatal toxicities in the DA1GO group, the DSMC recommended From www.bloodjournal.org by guest on May 8, 2018. For personal use only. randomization until relapse from CR or death from any cause, whichever closure of both the induction and postconsolidation randomizations. This occurred first, with observation censored at the day of last contact for patients recommendationwasreviewedandacceptedbythestudyteamandSWOG 4856 PETERSDORF et al BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 last known to be alive without report of relapse. OS was measured for all leadership, and the study was closed to accrual on August 20, 2009. patients from the day of initial randomization until death from any cause, with censoring at the day of last contact for patients last known to be alive. Relapse- Table 3. Treatment outcomes following induction chemotherapy of 595 adult patients with previously untreated AML, by treatment group free survival (RFS) was measured for patients who achieved CR from the day CR CR or Cri Resistant disease OS at 5 years RFS at 5 years Group Patients %95%CI%95%CI%95%CI%95%CI%95%CI of CR until relapse or death from any cause, with the same censoring as Results DA1GO 295 69 63-74 76 69-79 15 12-20From 46 www.bloodjournal.org 40-52 43 36-50 by guest on May 8, 2018. For personal use only. DFS. Adverse events were graded according to version 3.0 of the DA 300 70 64-75 74 69-79 20 16-25 50 44-56 42 35-49 Common Terminology Criteria for Adverse Events (accessible at http:// From AugustP* 2004 through August.59 2009, 637 adult patients.36 with .065 .85 .40 ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm). AML were*One-sided randomlyP value assigned for superior outcometo induction (higher CR with rate, lower the RD, DA HR,11)GO in DABLOOD,1 orGO group, 13 based JUNE on Fisher’s 2013 x exactVOLUME test (CR, 121, RD) NUMBERor logrank test 24 (OS, RFS). GEMTUZUMAB OZOGAMICIN IN AML INDUCTION/MAINTENANCE 4857 DA regimen. Thirty-nine patients (20 DA1GO and 19 DA) were 100 mg/m2 by continuous IV infusion on days 1 through 7. Marrow response CR/CRi/PR, and RD. Logrank tests and proportional hazards regression Statistical considerations was assessed on day 14, and if there was aplasia with less than 5% blasts, were used to analyze OS, RFS, and DFS. All HR values comparing treat- growth factor therapy (sargramostim (rhGM-CSF), filgrastim (G-CSF) or ment groups are for DA1GO relative to DA or GO relative to observation; group, at 15% vs 20%, but this difference was not statistically This study had 2 primary objectives: to test whether the CR rate was higher pegfilgrastim) could begin at the treating physician’s discretion. For both therefore, an HR lower than 1 indicates a superior outcome in the GO- significant (P 5 .065). among patients randomly assigned to the DA1GO group and to see groups, a second course using the DA regimen was allowed for patients with containing group. The statistical significance of treatment effects on marrow having more than 20% cellularity and more than 40% blasts on day response, OS, DFS, and RFS is represented by 1-sided P values for superior A total of 293 patients have died, and the remaining 302 were last whether postconsolidation DFS was higher among patients randomly 14, or with 5% or more blasts on a subsequent examination. The induction outcomes in the DA1GO induction or GO maintenance groups; all other known to be alive between 14 days and 7.1 years (median, 4.1 years; assigned to the GO group. A total of 342 evaluable patients were required for randomization was stratified by patient age (,35 years vs >35 years). P values are 2-sided. Confidence intervals (CIs) are at the 95% confidence the second objective. This number of patients, accrued over the course of 4.5 Patients who achieved CR and were afebrile and free of infection, with level. The following results were based on data available February 3, 2013. only 6 patients have less than 120 days’ follow-up). As shown in adequate organ function, performance status, and resolution of any central Figure 1, OS was not significantly better in the DA1GO group, with years and with 3 years’ additional follow-up, would ensure that a 1-sided test nervous system involvement, were eligible to receive 3 courses of consolidation Interim analyses and early closure at the 2.5% critical level had 90% power if the true DFS hazard ratio (HR; therapy with cytarabine 3 g/m2 by 3-hour continuous IV infusion every 12 a median of 41 months compared with 61 months in the DA group This study was monitored by the SWOG Data and Safety Monitoring hours on days 1, 3, and 5. Consolidation courses were administered monthly. GO: observation) is 0.67. This HR corresponds to increases in 1-year DFS Committee (DSMC). Interim analyses of induction results were scheduled (HR, 1.13; 95% CI, 0.90-1.42; P 5 .59). The trend toward longer OS After completing consolidation therapy, patients who continued to meet when 228 and 456 patients were evaluated for response to induction fi from 50% to 63% and from 75% to 83% for patients with unfavorable and the criteria for consolidation and who had not experienced sinusoidal ob- with DA was primarily a result of a signi cantly higher number of chemotherapy. Three interim analyses of postconsolidation DFS were favorable/intermediate cytogenetics, respectively. It was also predicted that structive syndrome during or after induction therapy were eligible for scheduled when 25%, 50%, and 75% of the expected number of events early deaths in the DA1GO group: 17 DA1GO patients died within postconsolidation randomization (1:1) between GO (5 mg/m2, 3 doses at half of all patients entering the study would enter the postconsolidation occurred, respectively. On August 11, 2009, the DSMC reviewed the least 28 days apart) vs observation. The postconsolidation randomization 30 days compared with only 4 DA patients. fi second scheduled interim analysis of CR rates, which was based as planned randomization; that is, that 684 patients would be available for the rst was stratified by prior use of GO (yes vs no) and preinduction cytogenetic fi on the first 456 evaluable patients. The CR rates in that analysis were 66% The effect of treatment on OS did not vary signi cantly among risk group (favorable vs intermediate vs unfavorable vs indeterminate). objective. This would ensure 90% power if the true CR rates were 81% with in 227 patients in the DA1GO group and 69% in 229 patients in the DA Patients were required to have an absolute neutrophil count higher than cytogenetic risk categories (P 5 .45). Among patients with favor- DA1GO and 70% with DA (2-sided test at the 5% critical level). group, and the hypothesis that the DA1GO regimen increases the CR 1000/mm3 and a platelet count higher than 100 000/mm3 to receive each fi rate by 12% was rejected at the predefined significance level (P , .0025). able cytogenetics, OS was somewhat, although not signi cantly, Data were collected and evaluated according to the standard practices of cycle of postconsolidation GO. Additional analyses showed that RFS was not significantly better on the DA1 better in the DA1GO group (HR, 0.54; 95% CI, 0.19-1.55; P 5 .12). SWOG. Fisher’s exact text and logistic regression analysis were used to fi Figure 1. OS of 595 adult patients with AML by induction treatment group. TickFigureGO group. 2. RFS The of DSMC 415 adult also reviewedpatients withthe rst AML planned who achieved interim analysis complete of response, However, in the 3 other cytogenetic risk groups, there was no analyze the effects of treatment group and other covariates on CR, CR/CRi, marks indicateTreatment censored outcomes observations. bypostconsolidation induction treatment DFS. group. That analysisTick marks rejected indicate the hypothesis censored observations. that GO fi fi CR, CR with incomplete hematologic recovery (CRi), partial response (PR), improves DFS, with a hazard ratio (observation: GO) of 1.5 at the prespeci ed suggestion of bene t with DA1GO (Table 4). In multivariate fi and resistant disease (RD) were defined according to the International Working signi cance level (P , .001). On the basis of these results, as well as the higher analysis, OS decreased significantly with increasing age and abso- 13 ineligibleincidence of because fatal toxicities of secondary in the DA1GO AML group, (n the5 DSMC22), recommended diagnosis other than Group Guidelines. DFS was measured from the day of postconsolidation Petersdorf S.H. et al Blood 2013 fi randomization until relapse from CR or death from any cause, whichever AMLclosure (n of5 both9), the acute induction promyelocytic and postconsolidation leukemia randomizations. (M3 AML) This (n 5 5), or lute peripheral blast count, was signi cantly poorer for patients occurred first, with observation censored at the day of last contact for patients recommendationwasreviewedandacceptedbythestudyteamandSWOG with performance status 2 to 3, and varied significantly among last known to be alive without report of relapse. OS was measured for all ageleadership, older and than the study 60 years, was closed inadequate to accrual on organ August 20, function, 2009. or coexisting patients from the day of initial randomization until death from any cause, with malignancy (n 5 1 each). Three additional patients (2 DA1GO and cytogenetic risk groups. Adjusting for these covariates, treatment censoring at the day of last contact for patients last known to be alive. Relapse- group had little effect on the treatment comparison (HR, 1.19; 95% free survival (RFS) was measured for patients who achieved CR from the day 1DA)refusedtoparticipateafterrandomization.Thefollowing of CR until relapse or death from any cause, with the same censoring as analysesResults are based on the remaining 595 patients (295 DA1GO and 300 CI, 0.94-1.53; P 5 .92). DFS. Adverse events were graded according to version 3.0 of the Of the 415 patients who achieved CR, 194 have relapsed and Common Terminology Criteria for Adverse Events (accessible at http:// DA).From Pretreatment August 2004 characteristicsthrough August of 2009, these 637 patients, adult patients including with karyotype, ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm). wereAML balanced were randomly between assigned the induction to induction groups with (Tables the DA1 1GO and or 2). another 33 have died with no report of relapse. RFS was slightly, DA regimen. Thirty-nine patients (20 DA1GO and 19 DA) were but not significantly, better in the DA1GO group (HR, 0.97; Statistical considerations Induction therapy 95% CI, 0.75-1.26; P 5 .40; Figure 2). Similar results were seen in This study had 2 primary objectives: to test whether the CR rate was higher the 445 patients who achieved CR or CRi (207 relapsed and 41 among patients randomly assigned to the DA1GO group and to see Five patients (3 DA1GO and 2 DA) received no protocol therapy whether postconsolidation DFS was higher among patients randomly others died; HR, 0.99; 95% CI, 0.78-1.28; P 5 .48). In multivariate assigned to the GO group. A total of 342 evaluable patients were required for because of refusal, physician’s recommendation, lack of insurance analysis, RFS decreased significantly with increasing age and the second objective. This number of patients, accrued over the course of 4.5 coverage, or injuries unrelated to treatment. In addition, 3 DA1GO years and with 3 years’ additional follow-up, would ensure that a 1-sided test white blood cell count. Adjusting for these covariates had little at the 2.5% critical level had 90% power if the true DFS hazard ratio (HR; patients did not receive GO because of elevated liver function tests, effect on the treatment comparison (HR, 0.98; 95% CI, 0.75-1.27; GO: observation) is 0.67. This HR corresponds to increases in 1-year DFS lung hemorrhage, and the trial’s closure, and 2 patients in the DA from 50% to 63% and from 75% to 83% for patients with unfavorable and P 5 .43). RFS was better in the DA1GO group among patients with favorable/intermediate cytogenetics, respectively. It was also predicted that group were treated with the DA1GO regimen. favorable cytogenetics (HR, 0.49; 95% CI, 0.21-1.18; P 5 .043; half of all patients entering the study would enter the postconsolidation Treatment outcomes are summarized in Table 3. The CR rate randomization; that is, that 684 patients would be available for the first Figure 3), but not in the other cytogenetic groups (Table 4). objective. This would ensure 90% power if the true CR rates were 81% with was 69% in the DA1GO group and 70% in the DA group (P 5 .59). DA1GO and 70% with DA (2-sided test at the 5% critical level). fi CR rates after the rst induction course were also similar (61% with Consolidation therapy Data were collected and evaluated according to the standard practices of DA1GO and 59% with DA), as were the CR rates of patients who SWOG. Fisher’s exact text and logistic regression analysis were used to Figure 1. OS of 595 adult patients with AML by induction treatment group. Tick analyze the effects of treatment group and other covariates on CR, CR/CRi, receivedmarks indicate a second censored inductionobservations. course of DA (44% for 54 patients in the Of the 415 patients who achieved CR, 376 were registered for DA1GO group and 48% for the 66 patients in the DA group). The consolidation, although 2 of these were ineligible for consolidation rates of CRi and PR in the 2 groups were 6% and 1% with DA1GO because of inadequate liver function. Of the 39 patients not registered, and 4% and 1% with DA. Including CRi and PR as responses did not 19 were removed to pursue hematopoietic cell transplantation alter the conclusion that response rates to DA1GO and DA alone (DA1GO 5 10; DA 5 9). Three of the 374 eligible patients were similar. The rate of RD was somewhat lower in the DA1GO received no protocol consolidation because of AML relapse, refusal,

Table 4. Induction treatment outcomes of 595 adult patients with AML by treatment group, within cytogenetic risk categories CR Resistant disease OS RFS Cytogenetic category and group Patients CR CR% P*RDsRD% P*DeathsHR†P*EventsHR† P*

Favorable DA1GO 37 29 78 .99 2 5 .91 5 0.54 .12 7 0.49 .043 DA 44 41 93 1 2 11 1.00 18 1.00 Intermediate DA1GO 137 103 75 .76 18 13 .26 68 1.24 .88 59 0.98 .46 DA 132 103 78 22 17 57 1.00 61 1.00 Unfavorable DA1GO 62 34 55 .26 15 24 .11 41 0.96 .42 19 0.90 .38 DA 55 26 47 20 36 39 1.00 17 1.00 Unknown DA1GO 59 39 66 .22 10 17 .15 37 1.25 .83 26 1.23 .76 DA 69 40 58 18 26 35 1.00 20 1.00

*One-sided P value for superior outcome (higher CR rate, lower RD, HR , 1) in DA1GO group, based on Fisher’s exact test (CR, RD) or Cox regression likelihood ratio 1test (OS, RFS). †HR, hazard ratio of DA1GO relative to DA. From www.bloodjournal.org by guest on May 8, 2018. For personal use only.

BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 GEMTUZUMAB OZOGAMICIN IN AML INDUCTION/MAINTENANCE 4859

Table 5. DFS of 169 adult patients with AML by postconsolidation Two additional randomized trials have shown benefit in the treatment group, within cytogenetic risk categories addition of GO to chemotherapy for older adult patients. In the DFS Acute Leukemia French Association 0701 trial, 280 patients aged Cytogenetic category and group Patients Events HR* P† 50 to 70 years were randomly assigned to receive 3 doses of GO at 2 Favorable 3 mg/m on days 1, 4, and 7 of DA induction, and again on day 1 16 GO 19 13 3.67 1.00 of each of 2 courses of consolidation therapy with DA. There Observation 21 5 1.00 was no significant difference in CR rate between the 2 groups. Intermediate However, there was improved event-free survival (40.8% vs 17.1% GO 53 33 1.03 .54 at 2 years; P 5 .0003), RFS (50.3% vs 22.7%; P 5 .0003), and OS Observation 52 31 1.00 (53.2 vs 41.9%; P 5 .037) for those patients who received Unfavorable GO. Survival benefit was seen in patients with favorable or GO 3 2 4.55 ND‡ intermediate-risk cytogenetics, but there was no benefit in patients Observation 4 1 1.00 with poor risk karyotype. There also was no difference in mortality Unknown GO 10 7 1.31 .67 between the 2 groups, but greater thrombocytopenia was seen in Observation 7 4 1.00 the GO treatment group. In the MRC AML16 trial, 1115 older patients were randomly assigned to receive 3 mg/m2 of GO on *HR, hazard ratio of DA1GO relative to DA. day 1 of either DA or daunorubicin plus clofarabine.17 There was †One-sided P value for superior outcome (HR , 1) in DA1GO group, based on fi Cox regression likelihood ratio test. no signi cant difference in CR rate or toxicity between groups. ‡P value not calculated because of small sample size. GO group: 2 relapses at 49 With a median follow-up of 30 months, the 3-year cumulative days and 15 months after postconsolidation randomization and 1 alive in CR at 42 incidence of relapse was significantly lower with GO (68% vs 76%; months; observation group: 1 relapse at 19 months and 3 alive in CR at 36, 58, and 66 fi months. P 5 .007), and 3-year survival was signi cantly better (25% vs 20%, P 5 .05). The benefitappearedtobepresentacrossdiseasesubgroups. group was 5.5% vs 1.4% in the standard group, with only a single S0106 failed to find a beneficial effect for the 3-drug com- death in the DA1GO resulting from liver toxicity. The remarkable bination of gemtuzumab ozogamicin, daunorubicin (at 45 mg/m2 result was the very low induction death rate of 1.4% in the control per day for 3 days), and cytarabine compared with daunorubicin group. The induction death rate in the DA1GO group was similar (at 60 mg/m2 per day for 3 days) and cytarabine alone. Why S0106 to that seen in contemporaneous large phase 3 trials with a similar failed to find a similar benefit as seen in the previously mentioned population. Eastern Cooperative Oncology Group 1900 had an trials is unclear. All the other trials administered GO on day 1 of induction death rate of between 4.5% and 5.5% in the 2 groups2; therapy instead of day 4, as was done in S0106. Most of the other Medical Research Council (MRC) AML 15, which randomly studies (with the exception of MRC 16) also exposed patients to assigned patients either to receive GO or not, had an induction GO during both induction and early in consolidation. Further, none rate of 7% (GO group) vs 6%.12 The induction toxicity rate seen in of the other trials attenuated the dose of anthracycline in the GO the GO group of this study was very typical for chemotherapy trials combination. Finally, the current trial was unusual, in that the for patients of this age; the very low death rate in the control group standard induction chemotherapy had less than a 2% induction accounted for the difference in the fatal induction rate for this trial. mortality rate, which is the lowest ever seen in a cooperative group In contrast to S0106, 4 phase 3 randomized trials have shown study. Any or all of these factors may have contributed to the potential benefit to the addition of GO to induction therapy in inconsistency in conclusions among trials. particular circumstances. The MRC AML 15 study included 1113 The role of GO in the treatment of AML remains poorly defined, patients who were randomly assigned to receive 3 mg/m2 of GO on and the optimal dose and schedule have not been established.18 day 1 of induction with DA, DA plus etoposide, or fludarabine, Given the rapid reexpression of the antigen after antibody binding, cytarabine, G-CSF, and idarubicin.14 Those randomly assigned to it is interesting to speculate whether nonhematopoietic toxicities GO also received the drug during consolidation. Although there was can be avoided and efficacy improved by using lower, repetitive no difference in overall CR rate, RFS, or OS among all patients, dosing, as was done in the Acute Leukemia French Association the effect of GO on OS varied significantly among cytogenetic risk 0701 trial. It is also uncertain how to identify patients most likely groups (P 5 .001). In particular, GO was associated with significantly better OS in patients with favorable risk karyotype (79% vs 51% at Table 6. Summary of induction toxicities among 586 adult patients 5 years; P 5 .0003), no benefit in those with adverse karyotypes, with AML fi and a trend for bene t in intermediate-risk patients. An internally DA1GO (n 5 292) DA (n 5 294) validated prognostic index identified approximately 70% of patients Patients % Patients % with a predicted benefit of 10% in 5-year survival. In the Groupe Ouest Est d’Etude des Leucemies´ Aigues¨ et Autres Maladies du Any fatal toxicity 16 5 4 1 Sang AML 2006 IR trial, 238 intermediate-risk patients aged 18 to Infection and/or febrile neutropenia 5 2 Central nervous system hemorrhage 4 1 60 years were randomly assigned to receive GO at 6 mg/m2 on Acute respiratory distress 30 day 1 with DA induction and during mitoxantrone plus cytarabine syndrome, dyspnea 15 consolidation. Patients with matched siblings were allocated Lung hemorrhage 2 0 to allogeneic hematopoietic cell transplantation. Considering all Transfusion related acute lung 10 patients, the 3-year event-free survival and OS in the GO group injury with infection and central were 51% and 53%, respectively, whereas in the group without nervous system hemorrhage GO, they were 33% and 46% (P 5 NS). In the subset that could Liver dysfunction 1 0 not receive an allogeneic transplant because they lacked a matched Other 0 1 sibling, there was improved event-free survival associated with the Any grade 41 nonhematologic 61 21 36 12 administration of GO (53.7% vs 27%; P 5 .0308). Any grade 31 nonhematologic 236 81 244 83

Petersdorf S.H. et al Blood 2013 2010

FDA DECIDE DI RITIRARE IL MYLOTARG DAL MERCATO LLM ADVANCES IN LLM

Current Developments in the Management of Leukemia, Lymphoma, and Myeloma • TREATMENT OF AML: RESURRECTION FOR GO? Section Editor: Susan O’Brien, MD

Gemtuzumab:VOLUME 30 ⅐ NUMBER Time 32 ⅐ NOVEMBER to 10Bring 2012 Back on the Market?

JOURNAL OF CLINICAL ONCOLOGY COMMENTS AND CONTROVERSIES James M. Foran, MD, FRCPC Associate Professor Division of Hematology & Medical Oncology GemtuzumabMayo Clinic Ozogamicin: Cancer Center Time to Resurrect? From www.bloodjournal.org by guest on May 11, 2018. For personal use only. Farhad Ravandi,Jacksonville,University of Texas MDFlorida Anderson Cancer Center, Houston, TX Elihu H. Estey and Frederick R. Appelbaum, Fred Hutchinson Cancer Research Center, Seattle, WA BloodFrancescoForum Lo-Coco, Tor Vergata University and Santa Lucia Foundation, Rome, Italy Charles A. Schiffer, Karmanos Cancer Institute, Wayne State University School of Medicine, Detroit, MI Richard A. Larson, University of Chicago Pritzker School of Medicine, Chicago, IL Alan K. Burnett, School of Medicine, Cardiff University, Cardiff, United Kingdom Hagop M. Kantarjian, University of Texas MD Anderson Cancer Center, Houston, TX GemtuzumabSee accompanying ozogamicin editorial on page 3905 in andacute article on myeloid page 3924 leukemia: a remarkable saga about an active drug On May 17, 2000, the US Food and Drug Administration (FDA) resulted in longer survival than was seen when treatment began at H&O What is gemtuzumab ozogamicin? toxicity9 concerns were reported. Excess liver toxicity was Jacob M. Rowegranted1,2 and accelerated Bob approval L¨owenberg for the3 use of gemtuzumab ozogamicin hematologic relapse. The relative rarity of APL, particularly of molec- (GO) in older patients (age Ն 60 years) with acute myeloid leukemia ular relapse, makesof a randomized particular study concern to confirm these in findings patients who went on to receive a 1Shaare Zedek Medical(AML) in Center, first relapse Jerusalem, who were Israel; not2Technion, considered Israel candidates Institute for of stan- Technology,infeasible. Haifa, OtherIsrael; areas and 3 inDepartment which GO of has Hematology, significant valueErasmus include University the JFMedical Gemtuzumab Center,dard Rotterdam, cytotoxic The chemotherapy. ozogamicin, Netherlands 1 Approval also for this known novel anti-CD33 by therelapsed trade AML and pediatricbone settings.marrow or stem cell transplant and in those who were immunoconjugate was based on a phase II trial demonstrating a 30% Although the beneficial results in APL alone would make the name Mylotargresponse rate (P (including!zer), completewas the response !rst [CR] monoclonal and CR with incom- antibodyavailability of GO desirable,receiving patients concurrent with APL constitute chemotherapy. only 10% to 2 approvedDespite living byplete in anthe platelet era US of recovery) unprecedented Foodand and was conditional Drugdrugs on Administration for future AML demonstration have received (FDA)12% regulatory of those with AML.withdrawal However,"ere of several this was phase agent also II trialswhen of an improved GO withEastern Cooperative Oncology of benefit in treatment of AML. Over the past 10 years, several phase II cytarabine plus anthracycline in relapsed AML have demonstrated progress in the understanding of the approval. In 2000, gemtuzumab ozoga- efficacy could not be demonstrated and in acute leukemia.and III trials have It addressed was granted this issue. accelerated approvalthat the for combination Group is safe and well (ECOG) tolerated,10,11 setting trial the stage called for E1900 that investigated genetic and molecularAreviewofthephaseIIstudiesin277patients(medianage,61 biology of acute micin (GO) was grantedits accelerated evaluation in newlytoxicity diagnosed appeared patients. excessive. Burnett et al12 Sincereported then, a patientsmyeloid leukemia agedyears) with60 (AML), relapsedyears this AML and has noted notolder a responseapproval in rate!rst of by 26%,relapse the essentially US Food of CD33- andrandomized Drug Ad- trial (MRCadministration4 randomized [Medical Research studies Council] of have gemtuzumabAML15) been show- com- right before autologous translated intoidentical significant to that in advances the studies that in led toministration the FDA approval. based3 GO on has promisinging that phase addition 2 ofpleted 3 mg/m that,2 GO toin cytarabine aggregate, and strongly daunorubicin sup- positivetherapy. Never acuteless GI before toxicity myeloid than have anthracyclines so leukemia many or cytarabinedata (AML) in (Ara-C), relapsed but who older distinc- adults wereinduction with not AML. in youngerstemport patients the cell (typically efficacy transplant age ofϽ this60 years) agent in with inyounger AML newly AML patients who were potential targetstively, been it has studied. been associated Yet most with hepaticGO sinusoidal held promise obstructive as a syn- new agentsignificantly that also improveddiagnosed survival in those AML with with cytogenetically acceptable favorable toxicity. considered candidates4 for cytotoxic chemotherapy. Gem- in !rst remission. No clinical bene!t was shown in the have not advanceddrome (SOS). beyondHowever, the phase the incidence 1 could of SOS be wasefficacious only 0.9% in in newlyrisk diagnosed and in 70% of patientsThere with is intermediate-risk a very plausible disease. explanation They devel- patients who did not undergo prior or subsequent allogeneic hemato- oped a model that reproducibly allowed identification of patients who tuzumaband, occasionally, targets phase CD33, 2 studies. which The AMLis present with acceptable on the toxicity. myeloid Several E1900for this discrepancy, trial, and making a British a compel- trial did not show an overall poietic cell transplantation (HCT).3 Although SOS was more frequent would live longer after treatment with the GO, anthracycline, and few ongoing phase 3 studies seem un- phase 3 studies were designed to test GO ling case for reapproval of GO in AML. blasts of approximatelyafter HCT, it was uncommon 85% if more of thanpatients 3.5 months with had elapsed AML.cytarabine "is combination. survival GO also improvedadvantage. outcomes when added to likely to havebetween more the lastthan GO a dose marginal and the HCT.in5 this setting. The results ofaregimenusinghigherdosesofcytarabine(FLAG-Ida[fludarabine, a random- (Blood. 2013;121(24):4838-4841) recombinant,benefit, if at all. Thus,Used humanized alone, it is GOnot is surprising most anti-CD33 effective inized acute promyelocyticstudy monoclonal by the leukemia Southwest antibodycytarabine, Oncology granulocyte colony-stimulating factor, and idarubicin]), attachesthat in past to few(APL), the decades likely cytotoxic because almost of a no highantitumor new surface expressionGroup antibiotic led of CD33, in 2010 the targetcalicheami to thesuggesting voluntary- that it was not merely a substitute for high-dose cytara- of GO. The combination of all-trans-retinoic acid (ATRA) and GO bine.12 However, theH&O randomized Can SWOG you 106 study, please which added discuss 6 the Acute Leukemia cin.Introduction In thiscan conjugate, be a substitute for the ATRA antibody plus anthracyclines binds in curing to newlyand ismg/m inter2 to- 3 ϩ 7(3-dayanthracyclineplus7-daycytarabine)inun-French Association (ALFA 0701) study, which diagnosed APL, producing a response rate of 84%,6 plausibly with less treated patients age Ͻ 60 years, found an increase in 30-day mortality nalizedGemtuzumab byacute ozogamicin tumor toxicity, less (GO)cells early consists andexpressing delayed of a cardiotoxicity, humanized the CD33 anti-CD33 and a lower antigen, riskstandard of uncompensated inductionand therapy bywas improvements with presented daunorubicin in CR, event-free and at cytarabine. survival the (EFS), In2011 this American Society of monoclonal antibodytherapy-related conjugated myelodysplastic with calicheamicin, syndrome–AML. a potent Hence, anti- as demon-trial, 637disease-free adult patients survival, younger or overall than survival 60 years (OS). were13 These randomly results assigned appar- thetumor attached anthracyclinestrated calicheamicin antib by Brecciaiotic. et The al,7 storyGO is of anis the attractive directly clinical option development fordelivered the treatment of toto of receive CD33-ently daunorubicin prompted Pfizer (45Hematology to mg/m voluntarily2 on days withdraw 1, 2,(ASH) GO 3) and from cytarabine the meeting? market (100 at expressingthis therapeuticolder agenttumor patients has been cells. with remarkable. APL, Calicheamicin with allGO treated is an active patients causes therapeutic responding double-strand withmg/m2theper request day by of continuous the FDA in infusion June 2010, on days before 1-7) the and results GO of (6 other mg/m2 durable molecular remissions. Furthermore, GO has been successfully randomized trials were available. agent but is not available on the market in the United States or Europe. on day 4) vs standard induction therapy with daunorubicin (60 mg/m2 DNA breakscombined and with inhibits ATRA and DNA arsenic trioxide synthesis. in newly diagnosed It is a pa-clever wayBecause of the pivotalJF In role the played ALFA by the SWOG trial, results, 280 com- patients with newly diagnosed In May 2000, GOtients, received particularly accelerated those with approval high-risk by disease, the US where Food the and high pre-on daysment 1-3) on and this study cytarabine seems appropriate. alone by continuous First, as in the infusion MRC study, on theredays 1 ofDrug targeting Administrationsenting a toxin WBC for treatment count to putsparticular as patients a single at increasedagent cells. of risk patients of early older death andthroughwas 7 a (DA). survival After advantage induction,AML for patients who patients with were were favorable to50–70 receive cytogenetics 3 years courses who of age underwent induction, as 2 than 60 years withrelapse. acute In a myeloid study conducted leukemia at the (AML) MD Anderson in first relapse Cancer who Center, theconsolidationreceived GO.therapy However, with unlike cytarabine in the MRC (3 g/m study,on the days doses 1, of 3, dauno- and 5). were not candidatesCR rate for was aggressive 81% in high-risk chemotherapy. patients who The received initial GO. approval8 The combi-After completionrubicin were of not the therapy,well same there inas the ! was tworst randomizedan and additional second arms. randomization In particular,consolidation to with daunorubicin plus was based on thenation results of ATRA of a phaseand arsenic 2 study trioxide of 142 plus patients GO is now with being AML evaluatedtest theto role arrive of atGO equitoxic as maintenance doses, the th dailyerapy daunorubicin vs observation. doses were The CRdesig- rate H&O Why was gemtuzumab removed from the cytarabine. Half2 of the patients were randomly assigned to also in first relapse.in The a North complete American respon Intergroupse (CR) APL rate trial for (SWOG all patients [Southwest was On-was 69%nated, for somewhatDA1GO andarbitrarily, 70% for as DA. 45 mg/m Similarly,in the the GO overall arm andefficacy, 60 cology Group] 0535) for high-risk APL. Italian investigators noted mg/m2 in the non-GO arm. The similar efficacy in both arms of the 16%, and when a subset of patients who had incomplete platelet as measured by the relapse-freereceive survival gemtuzumab, and the overall survivalwhich (OS), was added to the chemotherapy US marketthat earlyin treatment2010? of molecular relapse of APL with single-agent GO trial thus might be taken as evidence that GO contributed to this recovery (CRp) was added, the overall response rate was 30%. For was similar in both groups.at Theeach effect stage of treatment of treatment. on OS did not varyGemtuzumab was administered 1 fi patients older thanJournal 60 of Clinical years, Oncology, the overallVol 30, No response32 (November rate 10), 2012: was pp 26%. 3921-3923As signi cantly among cytogenetic© 2012 risk by American categories Society of(P Clinical5 .45), Oncology although3921 JFpart ofGemtuzumab the marketing approval, was the US voluntarily Food and Drug Administration removed among from patients the with favorableat a lowered cytogenetics, dose OS was schedule, not significantly in hopes that the toxicities of the marketmandated, fibyrst, completionP!zer in of the2010 ongoing after studies the of GOpostapproval in relapsed better trial in thethat DA 1GO groupdrug (P could5 .12). Therebe reduced. was no suggestion "e fractionated of dosing schedule of AML and, second, the initiation of randomized clinical trials com- benefit in the other cytogenetic risk groups. There was, however, 2 intendedparing GO in combination to provide with conv conentional!rmatory induction chemotherapy evidence ofincreased clinical induction mortalitygemtuzumab in the DA1GO consisted group, at 5% of vs 3 1% mg/m in on days 1, 4, and 7. to conventional chemotherapy alone.2 Final results of 3 phase 2 studies the DA group. On the basis of the results of these data, demonstrating beneof GO! ast adid single not agent inmeet relapsed its AML endpoints. confirmed the data"e from not the yetalackofclinicalbene formally fit and an" increasede addition mortality of in the gemtuzumab group of signi!cantly improved publishedearlier single phase S0106 2 study trial,1 and yieldedconducted a response by rate the of 13%Southwest CR patients Oncol who received- both GO, approval event-free for this and drug wasoverall voluntarily survival. "e number of patients and 13% CRp, for an overall response of 26%.3 withdrawn in June 2010. ogy Group (SWOG), was designed to determine Inwhether this study areexperiencing 2 factors to consider. events First, during the dose 3 years of of follow-up was lower by 2 addingThe SWOG gemtuzumab study in the current to issue standard of Blood chemotherapydaunorubicin demon- in the studyapproximately group, DA1GO, 25% was only in 45 the mg/m gemtuzumab arm (76 vs 104), compared with 60 mg/m2 in the standard group. Intensifying stratedIn this issue improved of Blood,4 thesurvival Southwest in Oncology previously Group (SWOG)untreatedanthracycline de novo doses inand induction the hasmedian been shown overall to have survival a sur- was 34 months versus 19.2 AMLdescribe patients the results of ages a phase 18–60 3 study, S0106,years. which "e was trial designed was to stoppedvival advantage, early particularlymonths, in younger respectively patients with AML.(P=.046).5,6 Thus, Toxicity was acceptable. "ere whencompare no in a prospectiveimprovement randomized in trial clinical the effects bene of adding!t was GO to observedthe 2 groups and may not bewere strictly 3 comparable, cases of and veno-occlusive the fact that the disease in the gemtuzumab

Submitted March 13, 2013; accepted April 11, 2013. Prepublished online as The publication costs of this article were defrayed in part by page charge Blood First Edition paper, April 16, 2013; DOI 10.1182/blood-2013-03-490482. payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734. 326 Clinical Advances in Hematology & Oncology Volume 10, Issue 5 May 2012 There is an Inside Blood commentary on this article in this issue. © 2013 by The American Society of Hematology

4838 BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 VOLUME 32 ⅐ NUMBER 27 ⅐ SEPTEMBER 20 2014

JOURNAL OF CLINICAL ONCOLOGY ORIGINAL REPORT

Gemtuzumab Ozogamicin in Children and Adolescents With De Novo Acute Myeloid Leukemia Improves Event-Free Survival by Reducing Relapse Risk: Results From the Randomized Phase III Children’s Oncology Group Trial AAML0531

Alan S. Gamis, Todd A. Alonzo, Soheil Meshinchi, Lillian Sung, RobertGamis etB. al Gerbing, Susana C. Raimondi, Betsy A. Hirsch, Samir B. Kahwash, Amy Heerema-McKenney, Laura Winter, Kathleen Glick, Stella M. Davies, Alan S. Gamis, Children’s Mercy Hospitals Patti Byron, Franklin O. Smith, and Richard Aplenc and Clinics, Kansas City, MO; Todd A. directly analyze the affect of SCT is restricted to IR AML. Fewer Alonzo, University of Southern California, Table 1. COG AAML0531 TherapeuticABSTRACT Regimen No-GO patients (45of 62 patients) received SCT as assigned than did Los Angeles; Robert B. Gerbing, Children’s Course and Agent Dose Days Oncology Group, Monrovia; Amy Heerema- GO recipeients (48 of 53 patients; P ϭ .015), primarily because of IND1Purpose McKenney, Stanford University, Palo Alto, To improve survival rates in children with acute myeloiddonor leukemia availability. (AML), Intent-to-treat we evaluated analysis (Appendix Table A3; Ap- CA; Soheil Meshinchi, Fred Hutchinson Cytarabine 100 mg/m2/dose twice per day IV 1 to 10 pendix Fig A1A) showed significantly improved DFS (Pϭ.02) and OS gemtuzumab-ozogamicin (GO),2 a humanized immunoconjugate targeted against CD33, as an Cancer Center; Laura Winter, Children’s Daunomycin 50 mg/m /dose IV 1, 3, 5 (P ϭ .02) with SCT. This benefit was limited to GO recipients and, Hospital and Regional Medical Center, alternativeEtoposide to further 100chemotherapy mg/m2/dose IV dose escalation. 1 to Our5 primary objective was to determine conversely, GO only benefited those patients who received SCT. Seattle, WA; Susana C. Raimondi, St Jude whetherGemtuzumab, adding arm B GO only to 3 mg/m standard2/dose IV chemotherapy over 2 hours improved 6 event-free survival (EFS) and overall Children’s Research Hospital, Memphis, IND2survival (OS) in children with newly diagnosed AML. Our secondary objectives examined TN; Betsy A. Hirsch, University of Minne- outcomesCytarabine by risk group 100 and mg/m method2/dose twice of perintensification. day IV 1 to 8 Univariable and Multivariable Analyses sota, Minneapolis, MN; Samir B. Kahwash, Daunomycin 50 mg/m2/dose IV 1, 3, 5 Nationwide Children’s Hospital, Columbus; Patients and Methods Risk factors found to be significant in univariable analysis (Ap- 2 Stella M. Davies, Cincinnati Children’s Children,Etoposide adolescents, 100 and mg/m young/dose adults IV ages 0 to 1 29 to 5 years withpendix newly Tables diagnosed A4 and A5) AML were includedwere in multivariable models to Hospital Medical Center; Franklin O. Smith, INT1enrolled onto Children’s Oncology Group trial AAML0531 and then were randomly assigned to 2 better define the impact of GO (Table 4). In multivariable analyses University of Cincinnati, Cincinnati, OH; eitherCytarabine standard five-course 1,000 mg/m chemotherapy/dose twice per alone day IV or to 1 to the 5 same chemotherapy with two doses of 2 adjusted for age, diagnostic WBC, race, and risk group, GO was inde- Kathleen Glick, Maine Children’s Cancer GOEtoposide (3 mg/m2/dose) administered 150 mg/m /dose once IV in induction 1course to 5 1 and once in intensification course 2 Program, Scarborough, ME; Richard pendently associated with better EFS (HzR, 0.80; 95% CI, 0.67 to 0.96; For(two patients of three). not undergoing Aplenc, Children’s Hospital of Philadelphia, stem-cell transplantation P ϭ .02), DFS (HzR, 0.80; 95% CI, 0.64 to 0.99; P ϭ .04), and RR Philadelphia, PA; Lillian Sung, Hospital for ResultsINT2 (HzR, 0.72; 95% CI, 0.57 to 0.91; P ϭ .006), as well as higher TRM Sick Children, Toronto; and Patti Byron, ThereMitoxantrone were 1,022 evaluable 12 mg/m patients2/dose IV enrolled. GO significantly 3 to 6 improved EFS (3 years: 53.1% v 46.9%; British Columbia Children’s Hospital, (HzR, 1.84; 95% CI, 0.97 to 3.47; P ϭ .06). Cytarabine 1,000 mg/m2/dose twice per day IV 1 to 4 Vancouver, Canada. hazard ratio [HzR], 0.83; 95% CI, 0.70 to 0.99; P ϭ .04) but not OS (3 years: 69.4% v 65.4%; HzR, 0.91; 95%Gemtuzumab, CI, 0.74 armto 1.13; B 3P mg/mϭ 2.39)./dose Although IV over 2 hours remission 7 was not improved (88% v 85%; P ϭ .15), Published online ahead of print at Toxicity posthoconly analyses found relapse risk (RR) was significantly reduced among GO recipients overall (3 www.jco.org on August 4, 2014. INT3 Common Terminology Criteria for Adverse Events v4 grade 3 to years: 32.8% v 41.3%; HzR,2 0.73; 95% CI, 0.58 to 0.91; P ϭ .006). Despite an increased Supported by Chair’s Grant No. U10 Cytarabine 3,000 mg/m /dose twice per day IV 1, 2, 8, 9 5toxicitiesweresimilarbetweenstudyarms(AppendixTableA6). postremission toxic mortality (32 years: 6.6% v 4.1%; HzR, 1.69; 95% CI, 0.93 to 3.08; P ϭ .09), CA98543-08 and Statistics and Data Center Escherichia coli L- 6,000 mg/m /dose IM 2, 9 Life-threatening sinusoidal obstruction syndrome (SOS) was similar Grant No. CA98413-08 of the Children’s disease-freeasparaginase survival was better among GO recipients (3 years: 60.6% v 54.7%; HzR, 0.82; 95% Oncology Group (COG) from the National ForCI, patients 0.67 to receiving 1.02; P ϭ .07). with one event in the No-GO arm during IND1, during SCT (No-GO Cancer Institute, National Institutes of matched family-donor v GO: two of 76 patients v three of 82 patients; P ϭ not significant Conclusion Health. stem-cell GOtransplantation added to chemotherapy improved EFS through a reduction in[NS]), RR foras was children SOS of and any degree adolescents (14 of 511 patients v18 of 511 patients; Presented in part at the 55th Annual Meet- withBusulfan, AML. 16 total doses Age and weight based Ϫ9 P ϭ NS). Acute left-ventricular systolic dysfunction was equivalent in ing of the American Society of Hematol- Ͻ 10 kg or Ͼ 4 years 0.8 mg/kg/dose once every 6 hours both arms (4.9% Ϯ 1.9% v 4.0% Ϯ 1.8%; P ϭ NS). Hematologic ogy, New Orleans, LA, December 7-10, old IV 2013. J Clin Oncol 32:3021-3032. © 2014 by American Society of Clinicaltoxicity Oncology was similar between study arms, including median time to Ͼ 10 kg and Ͻ 4 years 1 mg/kg/dose every 6 hours IV neutrophil recovery, which was more than 500/uL. However, posthoc Authors’ disclosures of potential conflicts old 3,4 of interest and author contributions are All patients Adjusted AUC based on first dose Ϫ27%8 to Ϫ to6 42%.analysisMatched to examine family-donor causes for (MFD)TRM differences found a higher pro- INTRODUCTION found at the end of this article. Cyclophosphamide 50 mg/kg/dose IV once per day Ϫtransplantation5 to Ϫ2 portion improved of GO disease-free patients duringsurvival INT2 rates with prolonged (Ͼ 59 days) Clinical trial information: NCT00372593. Abbreviations:Acute myeloid AUC, leukemia area under (AML)the concentration-time is among the curve; COG,(DFS) Chil- by betweenneutrophil 8% and recovery 10% and times postremission (12.0% v 6.3%; P ϭ .01). Corresponding author: Alan S. Gamis, MD, dren’smost Oncology difficult Group; to treat IM, of intramuscular; the childhood IND1, cancers induction course;overall INT, survival (OS)Though by between therapy 5% reductions and 13% in occurred two in similar proportions be- intensification course; IV, intravenous. MPH, Children’s Mercy Hospitals and Clin- because of its disease heterogeneity, high relapse, previous phasetween III arms trials.Gamis A. S. et al JCO 2014 (Appendix4,5 However, Table treatment- A6), death in remission was qualitatively ics, 2401 Gillham Rd, Kansas City, MO 1,2 higher among GO recipients (4.2% v 2.6%; P ϭ .21). Cumulative TRM 64108; e-mail: [email protected]. and toxic mortality. Therapeutic advances have related mortality (TRM) increased substantially included chemotherapy intensification and add- with therapyfrom intensification. enrollmentSupportive through last care follow-up im- without relapse or induction ©2014byAmericanSocietyofClinical OS was not improved (HzR, 0.91; 95% CI, 0.74 to 1.13; P ϭ .39; 3-year failure was higher in GO recipients (5-year4 TRM: GO, 8.6% Ϯ 2.5% v Oncology ing allogeneic stem-cell transplantation (SCT). provements reduced TRM (from 19% to 12%). OS: 69.4% Ϯ 4.2% v 65.4% Ϯ 4.4%). By risk group (Figs 2B to 2D), No-GO, 5.9% Ϯ 2.1%; P ϭ .09). This difference was primarily limited to 0732-183X/14/3227w-3021w/$20.00 Children’s Oncology Group (COG) legacy AML tri- However, it is increasingly evident that the limits of only EFS in the LR and IR groups suggested improvement with GO. the LR patients (two v eight patients;4,6,7 P ϭ .02) during INT2 and INT 3 als evaluated time-intensive induction and observed treatment intensification have been reached, neces- DOI: 10.1200/JCO.2014.55.3628 No difference in EFS or OS was detected in the HR patients when (Appendix Table A6), among those patients 11 years old or older (eight of improvement in event-free survival rates (EFS) from sitating alternative approaches. analyzed from study entry. 10 patients). All but one non-SCT TRM event during intensification occurred before neutrophil recovery and primarily late in the course © 2014 by American Society of Clinical Oncology 3021 Postremission Outcomes (mean, 56 days; range, 17 to 93 days) and was infection-related. Day-100 Postremission analyses suggested consistent differences by arm (Ta- TRM rates for MFD and alternative-donor SCT patients were 1.8% (n ϭ Downloadedble 3; Figs 3Afrom to ascopubs.org 3D). DFS among by 31.196.66.250 all GO recipients on May suggested 8, 2018 from improve- 031.196.066.2502) and 10.9% (n ϭ 5), respectively, and were similar between arms. TRM mentCopyright overall © and 2018 by American risk group Society (P ofϭ Clinical.07). Oncology. Exploratory All rights analyses reserved.beyond day 100 was equivalent. demonstrated a significant decrease in RR overall (HzR, 0.73; 95% CI, 0.58 to 0.91; P ϭ .006; 3-year RR: 32.8% Ϯ 4.6% v 41.3% Ϯ 4.9%), with qualitatively similar improvements within each risk group. In HR pa- DISCUSSION tients, the FLT3-ITD HAR cohort was the only one to benefit from GO (Appendix Figs A1B to A1C). However, OS after induction in the entire Using the largest randomized pediatric de novo AML trial to date and cohort and in each risk group was not improved. This was partially be- the only pediatric randomized controlled trial that added GO to in- cause of a higher postinduction TRM for GO recipients, particularly for duction and intensification, we have shown that EFS is significantly LR patients. improved by a significant reduction in relapse. These findings are consistent with recent randomized controlled trials in adults21,22,33 Stem-Cell Transplantation and together strongly supports the need to pursue therapeutic options SCT was recommended for all patients with HR AML and for usinganti-CD33antibody-drugconjugatesaddedtotraditionalchem- patients with IR AML if a MFD was available. Thus, the ability to otherapy and allogeneic SCT.

3024 © 2014 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Downloaded from ascopubs.org by 31.196.66.250 on May 8, 2018 from 031.196.066.250 Copyright © 2018 American Society of Clinical Oncology. All rights reserved. Gamis et al

A B 1.0 1.0

0.8 0.8

0.6 0.6

OS; GO OS; low-risk GO Survival Survival 0.4 OS; No-GO 0.4 OS; low-risk No-GO (probability) (probability) EFS; GO EFS; low-risk GO EFS; No-GO EFS; low-risk No-GO 0.2 0.2 OS P = .39 OS P = .74 EFS P = .04 EFS P = .18

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 Time Since Study Entry (years) Time Since Study Entry (years) No. at risk No. at risk OS; No-GO 511 415 342 244 149 49 1 OS; LR No-GO 121 117 102 74 48 13 0 OS; GO 511 407 353 254 141 50 5 OS; LR GO 125 110 101 76 46 21 4 EFS; No-GO 511 309 246 177 105 38 1 EFS; LR No-GO 121 96 79 59 36 11 0 EFS; GO 511 329 263 198 106 36 2 EFS; LR GO 125 102 86 65 37 14 2 C D 1.0 1.0 OS; high-risk GO OS; high-risk No-GO 0.8 0.8 EFS; high-risk GO EFS; high-risk No-GO

0.6 0.6

Survival 0.4 Survival 0.4 (probability) (probability) OS; intermediate-risk GO 0.2 OS; intermediate-risk No-GO 0.2 EFS; intermediate-risk GO OS P = .19 OS P = .78 EFS; intermediate-risk No-GO EFS P = .09 EFS P = .96

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 Time Since Study Entry (years) Time Since Study Entry (years) No. at risk No. at risk OS: IR No-GO 302 235 197 139 85 33 1 OS: HR No-GO 88 63 43 31 16 3 0 OS: IR GO 305 244 208 148 84 28 1 OS: HR GO 81 53 44 30 11 1 0 EFS: IR No-GO 302 178 142 100 60 25 1 EFS: HR No-GO 88 35 25 18 9 2 0 EFS: IR GO 305 194 150 113 62 22 0 EFS: HR GO 81 33 27 20 7 0 0

Fig 2. Overall survival (OS) and event-free (EFS) survival rates from study entry by study arm. (A) All patients; (B) low-risk (LR) patients; (C) intermediate-risk (IR) patients; (D) high-risk (HR) patients. GO, gemtuzumab-ozogamicin arm; No-GO, did not receive gemtuzumab-ozogamicin (control arm). Median survival ratesGamis A. S. et al JCO 2014 for each group is listed in Appendix Table A7, where applicable. courses of therapy. These last two courses were associated with the Limitations of this trial include its ability to show a statistically most prolonged median times to neutrophil recovery and adding GO significant improvement by AML risk group. This is, and will increas- seems to have worsened this in a subset of patients. Recent MRC ingly be, a challenge and a result of expanding heterogeneity of AML reports showed no benefit with a fifth course of therapy.7,42 COG no with ever smaller cohorts of relevant biologic factors. Even in adults in longer includes the final course of chemotherapy, which may lessen whom AML is much more prevalent, a five-trial meta-analysis was this risk in future GO trials. Also, the use of GO after remission may needed for adequate statistical power to determine GO’s impact on not be beneficial as seen in the NOPHO (Nordic Society of Pediatric outcome.36 Nevertheless, this is the largest pediatric AML trial re- Hematology and Oncology) trial.43 ported and likely represents the strongest evidence possible in a pedi- Although early GO studies saw increased SOS,44 we did not experi- atric randomized clinical trial. ence this. This is likely a result of our 3 mg/m2 GO dose selection and Our exploratory analyses determining reasons for a postin- timing, as GO doses of Ն6mg/m2 or SCT received within 120 days of GO duction improvement in DFS are admittedly posthoc. However, administration primarily increased this risk.44 Overall, toxicity during rather than a broad net of possible factors, this posthoc analysis SCT was not significantly greater in the GO arm. Acute cardiotoxicity, a focused on those associations that have repeatedly been found concern that affected SWOG’s choice of anthracycline dosing, was not in recent trials of adult patients. Our findings are consistent increased in our trial (although long-term observation is ongoing). De- with other GO trials and further strengthen the accumulated spite a higher infection-related TRM that attenuated GO’s affect on DFS literature. A new finding from our exploratory analyses was that and OS in our study, TRM observed in this trial compares favorably with the benefit of GO was limited to IR patients receiving SCT. This recent COG trials (Appendix Fig A2).3-5,16 association was further consistent with our finding that HR

3028 © 2014 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

Downloaded from ascopubs.org by 31.196.66.250 on May 8, 2018 from 031.196.066.250 Copyright © 2018 American Society of Clinical Oncology. All rights reserved. Gemtuzumab Ozogamicin COG Phase III Trial in Pediatric AML

A B Low-risk No-GO v GO, P = .77 1.0 Low-risk No-GO v GO, P = .38 1.0 Intermediate-risk No-GO v GO, P = .49 Intermediate-risk No-GO v GO, P = .24 High-risk No-GO v GO, P = .13 High-risk No-GO v GO, P = .16 0.8 0.8

0.6 0.6

0.4 0.4 Low-risk GO (proportion) (proportion) Low-risk No-GO

Low-risk GO High-risk GO Overall Survival Intermediate-risk GO 0.2 Low-risk No-GO High-risk No-GO 0.2 Intermediate-risk No-GO Disease-Free Survival Intermediate-risk GO High-risk GO Intermediate-risk No-GO High-risk No-GO

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 Time Since the End of IND2 (years) Time Since the End of IND2 (years) No. at risk No. at risk Low-risk No-GO 114 88 76 53 29 8 0 Low-risk No-GO 114 107 94 64 39 10 0 Low-risk GO 120 94 84 53 32 10 1 Low-risk GO 120 106 97 62 39 15 2 Intermediate-risk 257 156 127 93 50 22 1 Intermediate-risk 257 202 172 122 67 27 1 No-GO No-GO Intermediate-risk 268 179 142 97 55 18 0 Intermediate-risk 268 220 186 122 69 22 0 GO GO High-risk No-GO 47 27 20 13 4 1 0 High-risk No-GO 47 33 25 16 4 1 0 High-risk GO 41 29 23 16 5 0 0 High-risk GO 41 34 29 19 6 0 0 C D 1.0 Low-risk GO High-risk GO 0.30 Low-risk GO High-risk GO Low-risk No-GO High-risk No-GO Low-risk No-GO High-risk No-GO Intermediate-risk GO Intermediate-risk GO 0.8 Intermediate-risk No-GO 0.25 Intermediate-risk No-GO

Low-risk No-GO v GO, P = .04 0.20 0.6 Intermediate-risk No-GO v GO, P = .13 High-risk No-GO v GO, P = .08 0.15 Low-risk No-GO v GO, P = .04 0.4 Intermediate-risk No-GO v GO, P = .41

(proportion) (proportion) 0.10 Relapse Risk High-risk No-GO v GO, P = .65 0.2 0.05 Treatment-Related Mortality

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 Time Since the End of IND2 (years) Time Since the End of IND2 (years) No. at risk No. at risk Low-risk No-GO 114 88 76 53 29 8 0 Low-risk No-GO 114 88 76 53 29 8 0 Low-risk GO 120 94 84 53 32 10 1 Low-risk GO 120 94 84 53 32 10 1 Intermediate-risk 257 156 127 93 50 22 1 Intermediate-risk 257 156 127 93 50 22 1 No-GO No-GO Intermediate-risk 268 179 142 97 55 18 0 Intermediate-risk 268 179 142 97 55 18 0 GO GO High-risk No-GO 47 27 20 13 4 1 0 High-risk No-GO 47 27 20 13 4 1 0 High-risk GO 41 29 23 16 5 0 0 High-risk GO 41 29 23 16 5 0 0

Fig 3. Outcomes among patients from end of induction 2 (IND2) by risk group and study arm among patients in remission after the end of IND2. (A) Disease-free survival from end of IND2. (B) Overall survival from end of IND2. (C) Relapse risk from end of IND2. (D) Treatment-relatedGamis A. S. et al JCO 2014 mortality from end of IND2. GO, gemtuzumab-ozogamicin arm; No-GO, did not receive gemtuzumab-ozogamicin (control arm).

patients, all of whom received best available donor SCT, specif- AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS ically those who had FLT3-ITD HAR, also benefited from GO. OF INTEREST This will require validation in future trials, though is consistent with recent evidence that GO reduces minimal residual disease The author(s) indicated no potential conflicts of interest. and that reduced or absent minimal residual disease pre-SCT is associated with improved post-SCT DFS.45-47 Finally, our findings confirm CD33-targeted therapy added to AUTHOR CONTRIBUTIONS intensive chemotherapy improves EFS in de novo AML owing to a Conception and design: Alan S. Gamis, Todd A. Alonzo, Soheil Meshinchi, reduced relapse risk. As doses and schedules have varied among the 19,21,22 Laura Winter, Stella M. Davies, Franklin O. Smith, Richard Aplenc reported randomized trials, further investigation into optimal Administrative support: Kathleen Glick methods of GO administration and other CD33-targeted agents in Collection and assembly of data: Alan S. Gamis, Soheil Meshinchi, development should be pursued in future trials.10,48 Lillian Sung, Robert B. Gerbing, Susana C. Raimondi, Betsy A. Hirsch, www.jco.org © 2014 by American Society of Clinical Oncology 3029

Downloaded from ascopubs.org by 31.196.66.250 on May 8, 2018 from 031.196.066.250 Copyright © 2018 American Society of Clinical Oncology. All rights reserved. From www.bloodjournal.org by guest on May 8, 2018. For personal use only.

BLOOD, 13 JUNE 2013 x VOLUME 121, NUMBER 24 GEMTUZUMAB OZOGAMICIN IN ACUTE MYELOID LEUKEMIA 4839

Table 1. Randomized studies of GO in newly diagnosed patients with AML Dose of each Improved Improved RFS, Increased Increased administration CR with EFS, DFS or OS induction hepatic Study nAge,yearsCharacteristics of GO GO with GO mortality toxicity

SWOG 01064 637 18-60 DA1GO vs DA in 6mg No No Yes No induction and in maintenance MRC AML157 1113 ,60 Induction, 3mg No Yes:1.Favorable No No consolidation, cytogenetics 2. 70% and maintenance, of intermediate all with or without GO cytogenetics ALFA 07019,10 280 50-70 DA1GO vs DA in 3mg No Yes: In favorable/ No No induction and in intermediate group consolidation Groupe Ouest Est 254 18-60 Induction with or 6mg No Yes: Improved EFS No Yes d’Etude des Leuc´emies without GO Aigu¨eset Autres Maladies du Sang AML 2006 IR10 National Cancer 1115 51-84 Daunorubicin/ 3mg No Yes: In favorable/ No No Research Institute clofarabine induction, intermediate group AML168 with or without GO Leukemia Research Fund 495 Older adults, Low-dose cytarabine, 3mg Yes No No No AML14 and National for conventional with or without GO Cancer Research chemotherapy Institute AML 1611

CR, complete remission; DA, daunorubicin/cytarabine; DFS, disease-free survival; EFS, event-free survival; FS, relapse-free survival. efficacy was similar could suggest that the comparable response group, those 30% of patients predicted to not benefit were older and in the DA1GO group may actually be related to the dose had higher white blood cells counts at presentation and a worse intensification through to the additionofGO.Second,although performance status or secondaryRowe J. M. et al Blood 2013 disease.7 this was a randomized trial that should inherently account for The National Cancer Research Institute in Britain reported the differing populations, it is noteworthy that the mortality of 5% results of their large AML 16 study in 2012,8 in which 1115 patients in the DA1GO group is consistent with a typical induction with AML aged 51 to 84 years (median age, 67 years) were randomly mortality, as reported in the overwhelming majority of major assigned to receive daunorubicin/clofarabine with or without GO, studies in AML. In contrast, the very low induction mortality of 3mg/m2,onthefirst course of therapy. Once again, there was no dif- 1% in the control group is unprecedented. ference in the initial response rate, and the treatment-related mortality was similar, without any increase in toxicity with GO. However, at a fi Other studies of GO in newly diagnosed AML median follow-up of 30 months, the relapse rate was signi cantly lower with GO (68% vs 76%; P 5 .007), and the 3-year OS was significantly In addition to the SWOG study, there have been 4 major randomized better (25% vs 20%; P 5 .05). This is perhaps remarkable because the studies of the use of GO in newly diagnosed patients with AML with study evaluated a single dose of GO at 3 mg/m2,although,again,GO results that are sharply in contrast to those reported by the SWOG was used in combination with daunorubicin at 50 mg/m2 for 3 days.8 (Table 1). In 2011, the Medical Research Council in Britain (MRC) The Acute Leukemia French Association (ALFA) presented published the results of the AML 15 trial,7 in which 1113 patients who the results of their 0701 trial at the plenary session of the American were predominantly younger than 60 years were randomly assigned Society of Hematology annual meeting in 2011, and the final to receive a single dose of GO (3 mg/m2)onday1ofthefirst of 2 results were published in 2012.9 In this trial, conducted in patients induction courses with 1 of 3 induction randomized regimens. This with newly diagnosed AML, aged 50 to 70 years, patients were randomization, with or without the addition of GO, was continued randomly assigned in induction to daunorubicin/cytarabine with or throughout consolidation and maintenance. A predefined subgroup without GO (3 mg/m2, on days 1, 4, and 7). Patients in remission analysis revealed an improved OS at 5 years for patients with fa- received 2 additional courses of daunorubicin/cytarabine as con- vorable cytogenetics (79% vs 71%) but no benefit for patients with solidation, with or without GO, at the same dose of 3 mg/m2,on unfavorable cytogenetics. There was also an OS benefit for some day 1 of each cycle. Thus, in this particular study, GO was admin- patients with intermediate-risk disease, as indicated by an internally istered in combination with daunorubicin at 60 mg/m2 for 3 days validated index using cytogenetics, age, and performance status, and was given in a more intense regimen on 3 separate days in which predicted that approximately 70% of all intermediate-risk more than 1 cycle. Similar to the MRC study, there was no sig- patients would have a 10% improvement in 5-year OS if given GO nificant difference in the CR rate between the 2 groups or in the during induction. The 30% of intermediate-risk patients not likely to treatment-related mortality. However, the event-free survival was derive a benefit from GO were those who were older or had higher superior in the GO group, at 40.8% vs 17% in the control group white cell counts, a worse performance status, or secondary disease. (P 5 .0003). The OS was similarly better in the GO group, at It should be noted that this advantage of GO appeared in the context 53.2% vs 41.9% in the control group (P 5 .037). Once again, of remission induction chemotherapy based on daunorubicin at subgroup analysis indicated that the major benefit occurred in a dose level of 50 mg/m2 (3 days), which today we would probably patients in the favorable/intermediate group, but not among those consider suboptimal. Furthermore, of note, in the intermediate-risk with an unfavorable karyotype.9 Thus, this study reported a survival ALFA-0701: Study Design ● Randomized Open-label Phase 3

Induction 1st Consolidation 2nd Consolidation

Arm A DNR 60 mg/m2 D1 to D3 DNR 60 mg/m2 D1 DNR 60 mg/m2 D1, D2 AraC 200 mg/m2 D1 to AraC 1g/m2/12h D1 AraC 1g/m2/12h D1 D7 to D4 to D4

2nd course if BM blasts >10% at D15 DNR Randomization 60 mg/m2 D1, D2 CR or CRp AraC 1g/m2/12h D1 to D3

Arm B DNR 60 mg/m2 D1 to D3 DNR 60 mg/mIf CR 2 D1 DNR 60 mg/m2 D1, D2 AraC 200 mg/m2 D1 to D7 AraC 1g/m2/12h D1 to D4 AraC 1g/m2/12h D1 à D4 GO 3 mg/m2 D1, D4, D7 GO mg/m2 D1 GO 3mg/m2 D1

AraC, cytarabine; BM, bone marrow; CR, complete response; CRp, complete response with incomplete platelet recovery; D, day; DNR, daunorubicin; GO gemtuzumab ozogamicin (Mylotarg). Figure adapted from Castaigne S, et al. Abstract Presented at the 56th ASH Annual Meeng and Exposion; December 6-9, 2014; San Francisco, CA. ALFA-0701: Baseline Characteriscs Control Group GO Group All Paents Paents 139 139 278 61.7 62.8 62.2 Age in years, median (IQR) (57.4-65.5) (59.3-66.8) (58.5-66.3) Age ≥60 86 (62%) 100 (72%) 186 (67%) Men 61 (44%) 77 (55%) 138 (50%) ECOG performance status 0 54 (39%) 50 (36%) 104 (37%) 1 65 (47%) 75 (54%) 140 (50%) 2 17 (12%) 13 (9%) 30 (11%) 3 1 (<1%) 1 (<1%) 2 (<1%) Not available 2 (1%) 0 2 (<1%) White Blood Cell Count (x109 per L; 5.0 6.9 5.9 median, IQR) (1.9-26.7) (2.3-30.4) (2.1-29.1) 67.5 66.0 67.0 Platelet count (x109 per L; median, IQR) (36.3-125.5) (36.5-118.5) (36.0-122.0) 88% 92% 90% Percentage of CD33-expressing cells (median, IQR) (57-96) (67-97) (63-97)

GO, gentuzumab ozogamicin; IQR, interquarle range. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Baseline Characteriscs (cont’d)

Control Group GO Group All Patients Cytogeneticsa Favorable 6 (4%) 3 (2%) 9 (3%) Intermediate 91 (66%) 91 (66%) 182 (66%) Unfavorable 30 (22%) 28 (20%) 58 (21%) NPM1 statusa Mutated 48 (35%) 45 (32%) 93 (33%) Wild type 90 (65%) 91 (65%) 181 (65%) FTL3-ITD statusa Positive 27 (19) 22 (16%) 49 (18%) Negative 111 (80%) 115 (83%) 226 (65%) CEBPA statusa Mutated 8 (6%) 10 (7%) 18 (6%) Wildtype 119 (86%) 110 (79%) 229 (82%) Genotypea Favorable 24 (17%) 24 (17%) 48 (17%) Unfavorable 101 (73%) 95 (68%) 196 (71%)

GO, gentuzumab ozogamicin . a. Not shown: paents with informaon unavailable. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Outcomes Odds Ratio Control Group GO Group P Value (95% CI) All patients 139 139 1.46* CR + CRp 104 (75%) 113 (81%) 0.25 (0.82-2.59) CR 100 (72%) 102 (73%)

CRp 4 (3%) 11 (8%)

Induction courses

1 104 (75%) 113 (81%)

2 35 (25%) 25 (18%)

Death before induction 1 (<1%) 0

Death during induction 5 (4%) 9 (6%) Resistant disease 29 (21%) 17 (12%) (no CR or CRp)

CI, confidence interval; CR, complete remission; CRp, complete remission with incomplete platelet recovery; GO, gemtuzumab ozogamicin). Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Event-Free Survival 100 Hazard Ratio = 0.58 (0.43-0.78); p=0.0003 80

60

Gentuzumab 40 Ozogamicin

Event-Free Survival (%) 20 Control

0 0 6 12 18 24 30 36 42 48 Number at risk Time (months) Control 139 92 52 23 10 5 1 0 0 Mylotarg 139 101 75 46 32 18 10 3 0

Control Group (n=139) GO Group (n=139) Time (months; median, range) 9.7 (8.0-11.9) 15.6 (11.7-22.4) Estimated rate at 2 years (95% CI) 17.1 % (10.8-27.1) 40.8 (32.8-50.8)

GO, gentuzumab ozogamicin. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Event-Free Survival by Cytogenetic Status (Final Analyses)

Favorable/Intermediate Unfavorable

1.0 P=0.041 1.0 P=0.85

0.8 0.8

0.6 0.6 % EFS Gentuzumab % EFS 0.4 Ozogamicin 0.4

0.2 0.2 Gentuzumab Control Ozogamicin

Control 0.0 0.0

0 10 20 30 40 50 60 0 10 20 30 40 50 Months Months

Castaigne S, et al. Abstract Presented at the 56th ASH Annual Meeng and Exposion; December 6-9, 2014; San Francisco, CA. ALFA-0701: Relapse-Free Survival

100 Hazard Ratio = 0.52 (0.36-0.75); p=0.0003

80

60 Gemtuzumab Ozogamicin 40

20 Control Relapse-Free survival (%) Log-rank p=0-0003 0 0 6 12 18 24 30 36 Time (months) Number at risk Control 104 83 39 19 6 3 1 Mylotarg 113 101 68 41 29 16 8

Control Group (n=139) Mylotarg Group (n=139) Time (months; median, range) 11.4 (9.9 – 14.5) 28.1 (15.0-NR) Esmated rate at 2 years (95% CI) 22.7% (14.5-35.7) 50.3% (41.0-61.6)

Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Overall Survival

Primary Analyses1 Long-term Analyses2

100 p=0.0368 10 p=0.18

80 8

60 6 % OS Mylotarg Mylotarg 40 4 Overall survival Control 2 20 Control

0 0 0 6 12 18 24 30 36 42 48 0 10 20 30 40 50 60

Months from inclusion Months from inclusion

GO, gentuzumab ozogamicin. 1. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. 2. Castaigne S, et al. Abstract Presented at the 56th ASH Annual Meeng and Exposion; December 6-9, 2014; San Francisco, CA. ALFA-0701: Hematologic Toxicity Duration of Treatment-induced Cytopenia (Median Days) Control Group GO Group Point Difference* (95% P Value (n=139) (n=139) CI) Neutropenia (<0.5x109 ‑cells per L) Aer inducon 22 (18-27) 22 (20-26) -0.4 (-2.6 to -1.8) 0.68 Aer first consolidaon 10 (8-15) 13 (10-18) -2.9 (-5.4 to -0.6) 0.0017 Aer second consolidaon 13 (10-16) 15 (12-20) -3.7 (-6.2 to -1.4) 0.0021 Thrombocytopenia (<50 x 109 cells per L) Aer inducon 21 (18-25) 25 (20-30) -3.3 (-5.8 to -0.8) 0.0006 Aer first consolidaon 9 (6-13) 17 (11-27) -9.5 (-16.4 to -2.8) <0.0001 Aer second consolidaon 13 (9-20) 24 (15-35) -9.5 (-13.5 to -5.4) <0.0001 Persistent Thrombocytopenia (<50 x109 cells per L) By day 45 aer inducon 0/139 4/139 (3%) 0 (0 to 0.9) 0.125 By day 45 aer first 2/98 (2%) 9/99 (9%) 0.2(0.1 to 0.9) 0.05825 consolidaon By day 45 aer second 2/90 (2%) 9/85 (11%) 0.2 (0.1 to 0.8) 0.02895 consolidaon

Data are median (interquarle range [IQR]) or n/N (%) unless otherwise indicated. *All values mean difference except for persistent thrombocytopenia, which reflects relave risk. GO, gemtuzumab ozogamicin (Mylotarg). Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Non-Hematologic Toxicity Control Group Relave Risk GO Group (n=139) P Value (n=139) (95% CI) Inducon death 5/139 (4%) 9/139 (6%) 0.56 (0.20–1.54) 0.41 Transfer to intensive-care unit 17/139 (12%) 20/139 (14%) 0.85 (0.47–1.54) 0.72 Treatment-related death during 8/104* (8%) 2/113 (2%) 4.35 (1·.07–17.84) 0.051 CR or CRp Grade 3 and 4 AEs Hemorrhage 4/139 (3%) 12/139 (9%) 0.33 (0.12–0.95) 0.068 Cardiac 9/139 (6%) 11/139 (8%) 0.82 (0.36–1.87) 0.82 Liver 9/139 (6%) 18/139 (13%) 0.50 (0.24–1.05) 0.10 Skin or mucosa 25/139 (18%) 32/139 (23%) 0.11 (0.03–0.42) 0.37 Gastrointesnal 14/139 (10%) 22/139 (16%) 0.64 (0.34–1.18) 0.21 Pulmonary 16/139 (12%) 16/139 (12%) 1.00 (0.53–1.90) 1.00 Grade 3 and 4 Infecons During inducon 50/131 (38%) 59/129 (46%) 0.83 (0.62–1.11) 0.26 During first consolidaon 38/95 (40%) 48/97 (49%) 0.80 (0.59–1.11) 0.19 During second consolidaon 38/82 (46%) 38/81 (47%) 0.99 (0.71–1.37) 0.99

Data are n/N (%) unless otherwise indicated. AE, adverse event; CR, complete remission; CRp, complete remission with incomplete platelet recovery. *Includes 5 deaths aer stem cell transplants. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. ALFA-0701: Conclusions

• Fraconated doses of gemtuzumab ozogamicin added to standard chemotherapy improved clinical outcomes in paents aged 50–70 years with de novo AML – Significantly improved EFS (primary endpoint) in paents with favorable or intermediate cytogenecs – Improvement in OS in treatment arm containing fraconated dosing of gemtuzumab ozogamicin suggested in primary analysis, but OS not stascally significant at final, long-term analysis • 3-3-3 gemtuzumab ozogamicin regimen associated with an acceptable safety profile and allowed delivery of high cumulave dose of gemtuzumab ozogamicin without excess toxicity – Hematologic toxicity, parcularly persistent thrombocytopenia, more common in gemtuzumab ozogamicin-containing treatment arm than in the control arm – Gemtuzumab ozogamicin use not associated with an increase in the risk of death from toxicity or in the incidence of any grade 3 or 4 AE

AML, acute myeloid leukemia; EFS, event-free survival; OS, overall survival. Castaigne S et al. Lancet. 2012;379(9825):1508-1516. th Castaigne S, et al. Abstract Presented at the 56 ASH Annual Meeng and Exposion; December 6-9, 2014; San Francisco, CA. 38 Hills et al. (2014) Meta-Analysis

Objective

● Meta-Analysis of individual paent data from 5 trials in adults in which gemtuzumab ozogamicin was given in combinaon with standard inducon chemotherapy – Does gemtuzumab ozogamicin provide overall benefit with acceptable early mortality? – What is the opmum dose and dosing schedule?

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Study Design and Selecon of Datasets • Data from 3,325 paents – Enrolled in 1 of 5 randomized controlled trials of GO given with a first course of intensive inducon chemotherapy vs. intensive inducon chemotherapy alone • All paents ≥15 years old with newly diagnosed AML (de novo or secondary) or high-risk myelodysplasc syndrome • Trials involving less intensive inducon regimens (not administered to induce complete remission) and/or paents with acute promyelocyc leukemia were excluded • Relevant randomized controlled trials published up to May 1, 2013 were idenfied by a PubMed search using search terms “randomized” and “gemtuzumab” • Individual trialists were also contacted to confirm idenficaon of all relevant studies and collect individual paent data

Hills RK, et al. Lancet Oncol. 2014;15:986-996. 40 Hills Meta-Analysis: Outcomes • Primary endpoint: overall survival • Secondary endpoints: – Complete remission with or without complete peripheral count recovery – 30-day mortality – Relapse-Free survival – Relapse risk – Death in complete remission – Survival from complete remission – Survival censored at stem-cell transplantation • Endpoints defined in accordance with revised International Working Group criteria, except that peripheral count recovery not required for complete remission

Hills RK, et al. Lancet Oncol. 2014;15:986-996. 41 Hills Meta-Analysis: Studies Included Median Age Dose and Median No. Eligibility in Years Schedule Follow-up for Trial Patients Criteria (Range) Chemotherapy of GO Survival

3 mg/m² on days ALFA-0701 1, 4, and 7 of (Castaigne de novo AML; aged 62 24.1 months 278 50-70 years (50-70) DA (3+7) chemotherapy, up (IQR 15.7-32.8) et al, 2012) to 5 mg per dose

AML, either de MRC AML15 novo or secondary; DA (3+10, then 3+8), (Burnett 50 ADE (3+10+5, then 3 mg/m² on day 1 86.0 months 1,099 mostly aged <60 (15-71) of chemotherapy (IQR 76.6-99.4) et al, 2011) 3+8+5), or FLAG-Ida years

NCRI AML16 AML, either de (Burnett novo or secondary, DA (3+10, then 3+8) or daunorubicin (days et al, 2012) or high-risk MDS; 67 3 mg/m² on day 1 45.5 months 1,115 (51-84) 1, 3, and 5) plus of chemotherapy (IQR 34.3-57.6) mostly aged ≥60 clofarabine (days 1-5) years

SWOG S0106 (Petersdorf et de novo AML; aged 47 DA (3+7) plus G-CSF 6 mg/m² on day 4 55.2 months 595 18-60 years (18-60) or GM-CSF of chemotherapy (IQR 46.0-66.3) al, 2013)

GOELAMS AML 2006 IR (Delaunay et de novo AML, aged 50.5 6 mg/m² on day 4 39.3 months 238 18-60 years (18-60) DA (3+7) of chemotherapy (IQR 29.1-44.4) al, 2011)

GO = gemtuzumab ozogamicin; AML = acute myelocyc leukemia; DA = daunorubicin plus cytarabine; ADE = daunorubicin, cytarabine, and etoposide; FLAG-Ida = fludarabine, cytarabine, G-CSF, and idarubicin; G-CSF = granulocyte colony-smulang factor; GM-CSF = granulocyte-macrophage colony-smulang factor; IQR = interquarle range; IR = immediate release; MDS = myelodysplasc syndrome; MRC = Medical Research Council; NCRI = Naonal Cancer Research Instute; SWOG = Southwest Oncology Group. Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Paent Characteriscs • Ages ranged from 15 to 84 years (median 58 years) • Of 3,325 participants – 1,842 (55%) were male – 2927 (88%) had de novo disease – 285 (9%) had secondary disease – 113 (3%) had high-risk myelodysplastic syndrome • NPM1 mutation data available for 1,370 (41%) patients, of whom 398 (29%) had NPM1 mutation • Data for FLT3 internal tandem duplications available for 1,802 (54%) patients, of whom 354 (20%) had FLT3 internal tandem duplication mutations

Hills RK, et al. Lancet Oncol. 2014;15:986-996. 43 Hills Meta-Analysis: Effect of Gemtuzumab Ozogamicin on Relapse Events/Patients p o-e Variance OR (Cl*) Gemtuzumab No gemtuzumab value ozogamicin group ozogamicin group

3mg.m2 single dose MRC AML15 213/466 237/478 -15.5 112.3 0.87 (0.68-1.11) NCRI AML16 272/396 286/376 -32.7 137.8 0.79 (0.63-0.98) Subtotal 485/862 523/854 -48.2 250.2 0.85 (0.73-0.93) 0.002

Test for heterogeneity between trials χ2=0.6; p=0.4

3mg.m2 fractionated ALFA-0701 49/113 61/104 -15.7 26.2 0.55 (0.33-0.91) Subtotal 49/113 61/104 -15.7 26.2 0.55 (0.33-0.91) 0.002 6mg.m3 dose GOELAMS AML2006 IR 31/109 36/102 -4.3 16.7 0.77 (0.41-1.46) SWOG 0106 94/222 101/222 -3.7 46.7 0.92 (0.63-1.35) Subtotal 125/331 137/324 -8.0 63.4 0.88 (0.69-1.13) 0.3 Test for heterogeneity between trials χ2=0.4; p=0.5 Total 659/1306 721/1282 -71.9 339.8 0.81 (0.73-0.90) 0.0001 Test for heterogeneity (five trials) χ2=5.4; p=0.2 Test for heterogeneity between subtotals χ2=4.4; p=0.1

0.1 1.0 10.0

Favors Favors no gemtuzumab gemtuzumab ozogamicin ozogamicin

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Effect of Gemtuzumab Ozogamicin on Relapse-Free Survival Events/Patients p o-e Variance OR (Cl*) Gemtuzumab No gemtuzumab value ozogamicin group ozogamicin group

3mg.m2 single dose MRC AML15 282/466 314/478 -20.8 148.8 0.87 (0.70-1.07) NCRI AML16 625/396 321/376 -27.2 158.5 0.84 (0.69-1.03) Subtotal 607/862 635/854 -48.0 308.2 0.86 (0.77-0.96) 0.006

Test for heterogeneity between trials χ2=0.1; p=0.8

3mg.m2 fractionated ALFA-0701 51/113 69/104 -19.1 28.7 0.51 (0.32-0.83) Subtotal 51/113 69/104 -19.1 28.7 0.51 (0.36-0.74) 0.0004 6mg.m3 dose GOELAMS AML2006 IR 43/109 48/102 -4.9 22.7 0.81 (0.47-1.39) SWOG 0106 118/222 122/222 -2.5 59.9 0.96 (0.69-1.34) Subtotal 161/331 170/324 -7.4 82.6 0.91 (0.74-1.14) 0.4 Test for heterogeneity between trials χ2=0.5; p=0.5 Total 819/1306 874/1282 -74.4 419.5 0.84 (0.76-0.92) 0.0003 Test for heterogeneity (five trials) χ2=8.2; p=0.08 Test for heterogeneity between subtotals χ2=7.7; p=0.02

0.1 1.0 10.0

Favors Favors no gemtuzumab gemtuzumab ozogamicin ozogamicin

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Effect of Gemtuzumab Ozogamicin on Overall Survival Events/Patients

Gemtuzumab No gemtuzumab o-e Var OR (Cl*) p value ozogamicin group ozogamicin group

3mg.m2 single dose MRC AML15 326/548 348/551 -14.7 168.3 0.92 (0.75-1.12) B NCRI AML16 447/559 466/554 -31.1 226.8 0.87 (0.73-1.03) 100 Subtotal 773/1107 814/1105 -45.7 395.1 0.89 (0.81-0.98) 0.02 90 80 Test for heterogeneity between trials χ2=0.2; p=0.6

2 3mg.m fractionated 70 Difference 3.7% (SD 2.0) ALFA-0701 59/139 72/139 -11.8 32.1 0.69 (0.44-1.09) Subtotal 59/139 72/139 -11.8 32.1 0.69 (0.49-0.98) 0.04 60 Log-rank p=0.01 50 6mg.m3 dose GOELAMS AML2006 IR 41/119 54/119 -7.0 23.7 0.75 (0.44-1.27) 35.6% SWOG 0106 151/295 144/300 8.0 73.6 1.11 (0.83-1.50) 40 34.3% Subtotal 192/414 198/419 1.0 97.3 1.01 (0.83-1.23) 0.9 Overall Survival (%) 30 32.2% 20 30.6% Test for heterogeneity between trials χ2=2.9; p=0.09 Allocated to GO Total 1024/1660 1084/1663 -56.6 524.5 0.90 (0.82-0.98) 0.01 10 Allocated to no-GO 0 2 Test for heterogeneity (five trials) χ =6.7; p=0.2 0 1 2 3 4 5 6+ 2 Test for heterogeneity between subtotals χ =3.6; p=0.2 Years

0.1 1.0 10.0 Annual Event Years 1-5 Years 6+ Rates Favors Favors gemtuzumab no gemtuzumab GO 26.7% SD 0.8 3.5% SD 0.8 ozogamicin ozogamicin

No-GO 29.5% SD 0.9 5.2% SD 1.0

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Overall Survival by Cytogenec Status Events/Patients p o-e Variance OR (Cl*) Gemtuzumab No gemtuzumab value ozogamicin group ozogamicin group

Original coding Favourable 32/125 54/126 -14.3 20.5 0.50 (0.32-0.77) Intermediate 549/962 596/964 -44.2 284.4 0.86 (0.76-0.96) Adverse 223/261 227/256 3.1 110.6 1.03 (0.85-1.24) Subtotal 804/1348 877/1346 -55.4 415.5 0.88 (0.79-0.96) 0.007

Test for heterogeneity between subgroups χ2=9.6; p=0.008

Test for trend between subgroups χ2=7.8; p=0.005

Revised MRC coding12 Favourable 30/122 54/124 -15.5 20.6 0.47 (0.31-0.73) Intermediate 506/911 559/916 -45.3 264.6 0.84 (0.75-0.95) Adverse 260/299 258/284 1.2 127.6 0.99 (0.83-1.18) Subtotal 796/1332 871/1324 -61.9 412.8 0.86 (0.78-0.95) 0.002

Test for heterogeneity between subgroups χ2=10.1; p=0.006 Test for trend between subgroups χ2=7.7; p=0.006

0.1 1.0 10.0

Favors Favors no gemtuzumab gemtuzumab ozogamicin ozogamicin

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: Overall Survival by Cytogenetic Status

Favorable Intermediate Adverse

100 100 100 Difference 5.7% (SD 2.8) Difference 2.2% (SD 9.8) 90 90 Log-rank p=0.005 90 Log-rank p=0.9 77.5% 75.5% 80 80 80 70 70 70 60 60 60 50 50 40.7% 50 55.0% 54.8% 39.6% 40 40 40 Difference 20.7% (SD 6.5) Overall Survival (%) Overall Survival (%) 30 30 35.5% 33.9% Overall Survival (%) 30 Log-rank p=0.0006 9.1% 8.9% 20 20 20 10 Allocated to gemtuzumab ozogamicin 10 10 Allocated to no gemtuzumab ozogamicin 0 0 0 0 1 2 3 4 5 6+ 0 1 2 3 4 5 6+ 0 1 2 3 4 5 6+ Years Years Years 7.9% 6.7%

Annual Event Annual Event Annual Event Years 1-5 Years 6+ Years 1-5 Years 6+ Years 1-5 Years 6+ Rates Rates Rates Gemutuzumab Gemutuzumab Gemutuzumab 5.8% SD 1.1 2.3% SD 1.3 22.4% SD 1.0 2.7% SD 0.9 73.8% SD 4.6 2.4% SD 0.8 ozogamicin ozogamicin ozogamicin No-gemutuzumab No-gemutuzumab No-gemutuzumab 21.1% SD 14.1% SD 1.9 0.0% SD 0.0 26.2% SD 1.1 4.9% SD 1.3 76.7% SD 4.8 ozogamicin ozogamicin ozogamicin 10.5

Hills RK, et al. Lancet Oncol. 2014;15:986-996. Hills Meta-Analysis: No Effect on 30-Day Mortality, Death in Complete Remission, or Survival aer Remission • Nonsignificant increase in 30-day mortality with gemtuzumab ozogamicin • 30-day mortality was significantly greater for patients given gemtuzumab ozogamicin at 6 mg/m2 than for those given 3 mg/m2 (heterogeneity P = .03) • No significant difference between treatment groups with respect to deaths while in complete remission • None of the trial results suggested that patient deaths while in remission were increased among those receiving gemtuzumab ozogamicin • Reduction of relapse with addition of gemtuzumab ozogamicin led to significantly improved survival after achieving remission

Hills RK, et al. Lancet Oncol. 2014;15:986-996. 49 Conclusions • GO can be added safely to convenonal inducon therapy and significantly improves overall survival – 10% reducon in risk of death (P = .01 ) – 16% reducon in risk of relapse (P = .0003) • 30-day mortality – Lower with 3 mg/m2 vs. 6 mg/m2 – When SWOG S0106 excluded, GO not associated with increased 30-day mortality for remaining paents • Cytogenecs showed significant interacon with treatment – Survival benefit strongest in those with favorable cytogenec characteriscs – Survival benefit also evident in those with intermediate characteriscs – Paents with adverse cytogenec characteriscs did not benefit

Hills RK, et al. Lancet Oncol. 2014;15:986-996. 50 AML-19

Objective

● Sequential Phase 2-3 design ‒ Phase 2: determine which of 2 schedules of low-dose GO induction monotherapy was more promising to continue phase III comparison with BSC in the study population ‒ Phase 3: to compare GO to BSC in untreated AML in older patients unfit for intensive chemotherapy

● Accrued June 2004 through Dec 2006 ● Phase 2 results published 2012 in Br J Haematol1 ● Phase 3 results published 2016 in JCO

BSC, best supporve care; GO gemtuzumab ozogamicin; OS, overall survival 1. Amadori S, et al. Br J Haematol. 2010;149(3):376-82. 2. Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-19: Overall Study Design

Phase 2 Phase 3 ARM A Induction GO 3 mg/m2 D1 Eligible Paents GO 3 mg/m2 D3 Randomized GO 3 mg/m2 D5 Pick the best for Untreated AML in Phase 3 older paents ARM B unfit for intensive Induction chemotherapy GO 6 mg/m2 D1 GO BSC (N=56) GO 3 mg/m2 D8 (Best Arm) Randomized Randomized

ARM C Best supportive Care

AML, acute myeloid leukemia; BSC, best supporve care; D, day; GO, gemtuzumab ozogamicin. Amadori S, et al. Br J Haematol. 2010;149(3):376-82. AML-19: Overall Study Design

Phase 2 Phase 3 ARM A Induction GO 3 mg/m2 D1 Eligible Paents GO 3 mg/m2 D3 Randomized GO 3 mg/m2 D5 Pick the best for Untreated AML in Phase 3 older paents ARM B unfit for intensive Induction chemotherapy GO 6 mg/m2 D1 GO BSC (N=56) GO 3 mg/m2 D8 (Best Arm) Randomized Randomized

ARM C Best supportive Care

AML, acute myeloid leukemia; BSC, best supporve care; D, day; GO, gemtuzumab ozogamicin. Amadori S, et al. Br J Haematol. 2010;149(3):376-82. AML-19: Eligibility Criteria

Inclusion Exclusion

● Previously untreated patients ● Acute promyelocytic leukemia

● de novo or secondary AML ● Central nervous system leukemia

● Not considered candidates for ● Blast crisis of CML or AML intensive chemotherapy – All >75 years ● Concomitant malignant disease – 61-75 years with a WHO PS >2 or ● Severe cardiac or pulmonary unwilling to receive intensive chemotherapy dysfunction ● Active uncontrolled infection

● HIV positivity

AML, acute myeloid leukemia; CML, chronic myeloid leukemia; PS, performance status; WHO, World Health Organizaon. Amadori S, et al. Br J Haematol. 2010;149(3):376-82. AML-019: Summary of Clinical Response by Treatment Arm, Phase 2 Treatment Arm All Paents B [D1, 8] Response A [D1,3,5] (n=29) (n=56) (n=27) N (%) N(%) N (%) CR 11 (20) 6 (21) 5 (18) CRp 1 (2) 0 1 (4) PR 1 (2) 1 (3) 0 Arm B: Highest SD 15 (26) 4 (14) 11 (41) Rate of DnP; Met the Statistical PD 19 (34) 12 (41) 7 (26) Criteria to be Selected as the DnP 28 (50) 11 (38) 17 (63%) Preferred Regimen for Death Phase III 7 (12) 4 (14) 3 (11) (≤6 weeks) Comparison with Best Un-assessable 2 (4) 2 (7) 0 Supportive Care

CR, complete remission; CRp, complete remission without platelet recovery; DnP, disease non-progression; PR, paral remission; SD, stable disease; PD, progressive disease. Amadori S, et al. Br J Haematol. 2010;149(3):376-82. AML-019: Phase 3 Study Design

If CR/CRi/PR/SD GO Induction GO Untreated AML in (118) Continuation older paents unfit for intensive 1:1 chemotherapy Randomization (N=237) Best supportive care (HU as necessary) (n=119)

GO Schedules Induction 6 mg/m2 day 1 + 3 mg/m2 day 8 Continuation 2 mg/m2 monthly (max 8)

AML, acute myeloid leukemia; BSC, best supporve care; CR, complete remission; CRi, complete remission with incomplete hematologic recovery; GO gemtuzumab ozogamicin (Mylotarg); HU, hydroxyurea; PR, paral response; SD, stable disease Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-019: Overall Survival

100 GO (N=118) BSC (N=119) 90 Median, mo 4.9 3.6 80 HR (95%, CI) 0.69 (0.53 – 0.90) Log-rank P value 0.005 70 1-year rate 24.3% 9.7% 60 50 40

Survival (%) 30 20 10 0 (months) 0 6 12 18 24 30 O N Number of patients at risk: 113 118 53 28 11 5 GO 115 119 33 11 7 3 BSC

BSC, best supporve care; GO, gemtuzumab ozogamicin; N, number of paents; O, observed number of events Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] Overall Survival by Patient Subgroup Deaths/Paents Interacon GO BSC HR 95% CI Test Sex Male 57/57 71/73 0.90 0.63 – 1.28 P = .05 Female 56/61 44/46 0.53 0.35 – 0.79 CD33 expression <20% 9/10 13/14 1.52 0.65 – 3.58 20-80% 58/58 58/58 0.75 0.52 – 1.09 P = .05 >80% 44/48 44/47 0.49 0.32 – 0.76 Cytogenec Risk Favorable/Intermediate 54/59 45/45 0.55 0.37 – 0.82 Adverse 33/33 29/32 1.11 0.67 – 1.83 P = .08 Unknown 26/26 41/42 0.85 0.52 – 1.40 Selected subgroups only BSC, best supporve care; CI, confidence interval; GO, gemtuzumab ozogamicin ’ HR, hazard rao Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-19: Response to Gemtuzumab Ozogamicin Response, % Best Response Induction Response (n=111) at Any Time CR + CRi 24.3 27.0 CR 8.1 15.3 CRi 16.2 11.7 PR 6.3 5.4 SD 39.6 38.7 Progressive Disease 14.4 Induction Death 7.2 Not Evaluable 8.1

CR, complete remission; CRi, complete remission with incomplete platelet recovery; PR, paral response; SD, stable disease. Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-19: Adverse Events on Study

AEs, % (Safety Populaon) GO (N=111) BSC (N=114) All-grade AEs 87.3 90.4 Grade ≥3 AEs 61.2 67.5 Deaths Due to AEs 17.1 20.2 30-day Mortality, % 11 13.5 60-day Mortality, % 17.8 30.4

AE, adverse, event; BSC, best supporve care; d, day; GO gemtuzumab ozogamicin . Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-19: Non-Hematologic Toxicity

Maximal Grade ≥3 Adverse Events (AEs) Occurring in >5% of Patients

AEs, % (Safety GO (N=111) BSC (N=114) Population) Infection 35.1 34.3 Febrile neutropenia 18.0 23.7 Bleeding 12.6 12.3 Fatigue 11.7 21.0 Liver 7.2 6.1 Cardiac 6.3 14.0 Metabolic 3.6 6.1 Renal 3.6 4.4

AE, adverse events; BSC, best supporve care; GO gemtuzumab ozogamicin . Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] AML-19: Conclusions

• In older paents with newly diagnosed AML unsuitable for intensive chemotherapy, GO significantly improved OS compared with BSC • Subgroup analyses revealed interacons between baseline CD33 expression, sex, and cytogenec profile and treatment effect for OS – GO significantly improved OS compared with BSC • In paents with >80% CD33-posive blasts • In women • In paents with favorable/intermediate cytogenec risk profiles • No apparent increase in toxicity – Incidence of adverse events similar in both arms – Deaths due to AEs less common with gemtuzumab ozogamicin • Further development of GO in this area of high unmet medical need is warranted

AE, adverse event; BSC, supporve care; GO, gemtuzumab ozogamicin; OS, overall survival Amadori S, Suciu S, Selleslag D et al. J Clin Oncol. 2016 Jan 25. pii: JCO640060. [Epub ahead of print] 62 MyloFrance-2: Phase I/II study of fractionated doses of MYLOTARG with escalated doses of DNR and Ara-C as first AML salvage Consolidation* First Induction relap 2 courses MYLOTA MYLOTA MYLOTA se RG + RG + RG CR • Amsacrine CD3 2 3 mg/m² 3 mg/m² 3 mg/m² , 90 mg/m 3+ Day 1 Day 4 Day 7 CR Days 1–3 AML patie +1 +2 +3 p • Ara-C 1 g/ + + + m2/12 hr nts Ara-C Ara-C Ara-C 50– • MYLOTARGDays 1–3 3 100 mg/ 100 mg/ 200 mg/ mg/m2 70 m² Days m² Days m² Days Day 1 year 1–7 1–7 1–7 s old + + + DNR DNR DNR N=20 45 mg/m² 60 mg/m² 60 mg/m² Days 1–3 Days 1–3 Days 1–3

§ Primary objective § Secondary objectives • Determine optimal – Remission rate DNR and Ara-C doses to be combined with – RFS fractionated dosing of – OS MYLOTARG MyloFrance-2: Key eligibility criteria

Inclusion criteria Exclusion criteria § 50–70 years old § APL § CD33+ AML in first § Secondary AML relapse § ECOG PS ≤2 § Serum creatinine ≤2.0 mg/dl § ALT and AST levels <2× ULN MyloFrance-2: Baseline characteristics Characteristics N or % Number of patients, n 20 Median age (range), years 60 (50–70) Median duration of CR1 (range), months 10 (6–42) Cytogenetics Evaluable, n 19 Favourable, n 2 Intermediate risk, n 15 Poor risk, n 2 MyloFrance-2: Overall results by dose level Dose Dose Dose level 3 level 1 level 2 N 4 4 4 8 Responder 2 2 4 6 CR 2 1 4 4 CRp - - 2 PR - 1 - Failure 2 1 1 Early death - 1 1 Grade 3/4 fever 2 1 1 0 Grade 3/4 infection 1 3 3 4 Grade 3/4 liver - 1 - 1 toxicity DLT 0 1 0 Overall, the third dose level was considered as tolerable, with only 1 DLT observed at dose level 2 MyloFrance-2: Overall results

Main efficacy outcomes N=20 CR/CRp, n (%) 13 (65.0) Median CR duration 12 months Median OS 15 months

Main safety outcomes N=20 Median duration of neutropenia <500/µl 30 days Median duration of thrombocytopenia <50,000/µl 32 days Early deaths, n (%) 2 (10.0) Grade 3/4 fever, n (%) 5 (25.0) Grade 3/4 infection, n (%) 11 (55.0) Grade 3/4 liver toxicity, n (%) 2 (10.0)

No episodes of VOD MyloFrance-2: Conclusions

MF-2: MYLOTARG (3 mg/m2/Days 1, 4, 7) in combination with DNR (60mg/m2/d Days 1–3) and AraC (200 mg/m2/d Days 1–7)

Good hepatic Time to recovery of tolerance was neutrophils and platelets observed, NO VOD was longer than (4 patients with previously reported HSCT) The results of the trial suggest that this combination is associated with acceptable toxicity SETTEMBRE 2017 • FDA RE-APPROVAL OF GO FOR THE TREATMENT OF NEWLY-DIAGNOSED CD33+ AML IN ADULTS AND FOR THE TREATMENT OF RELAPSED OR REFRACTORY CD33+ AML IN ADULTS AND IN PEDIATRIC PATIENTS > 2 YRS. GO MAY BE USED IN COMBINATION WITH DAUNORUBICIN AND CYTARABINE FOR ADULTS WITH NEWLY-DIAGNOSED AML, OR AS A STAND-ALONE TREATMENT FOR CERTAIN ADULT AND PEDIATRIC PATIENTS

• LE DOSI DI MYLOTARG DA UTILIZZARE SONO QUELLE DEGLI STUDY ALFA0701 E DELL’AML-19

• WARNING PER LA TOSSICITA’ EPATICA CORE BINDING FACTOR (CBF)-LEUKEMIAS

• I BLASTI CON t(8;21) NELLA MAGGIOR PARTE DEI CASI NON ESPRIMONO LA Pgp (OVVERO IL PRODOTTO DEL GENE MDR1), PROBABILMENTE PER UNA SELETTIVA REPRESSIONE DEL PROMOTER DI MDR-1 DA PARTE DI AML1-ETO. DIVERSI STUDI HANNO DIMOSTRATO COME LA ESTRUSIONE DI GO DA PARTE DI Pgp POSSA CONDIZIONARE LA RISPOSTA AL MYLOTARG

• LE LEUKEMIA-INIZIATING CELLS DELLE CBF AML, A DIFFERENZA DI ALTRI TIPI DI AML, ORIGINEREBBERO DA PRECURSORI MIELOIDI EARLY-COMMITTED PIUTTOSTO CHE DA HSCs PIU’ IMMATURE E QUINDI SAREBBERO PIU’ SENSIBILI AL MYLOTARG A CAUSA DELLA PIU’ ELEVATA ESPRESSIONE DI CD33 Brief Report HematologyCorrespondence: Reports 2017; Michele volume Gottardi, 9:7028 Hematology, Department of Specialty Core Binding Factor (CBF) Acute Medicine, General Hospital “Santa Maria di Myeloid Leukemia (AML) patients Ca’ Foncello”, piazza Ospedale 1, 31100 experience treatment failure in the order of Treviso, Italy. Clinical and experimental 40-50%.1 As such, efforts have been made to Tel.: +39.0422.322221 - Fax: +39.0422.322542 1 1 efficacyMichele Gottardi, of gemtuzumabFederico Mosna, increase the dose intensity of first line E-mail: [email protected] ozogamicinSergio de Angeli, in2 core binding treatment over the standard BrieffactorCristina Report acutePapayannidis, myeloid3 Anna leukemia Candoni, 4 Daunorubicin/CytarabineHematologyCorrespondence: Reports 2017; Michele3+7 volumeinduction, Gottardi, 9:7028Key words: Core binding factor; acute Marino Clavio, 5 Cristina Tecchio, 6 under the assumption that better survival myeloid leukemia; gemtuzumab ozogamicin; Hematology, Department of Specialtyimmunotherapy; autologous stem cell trans- Andrea Piccin,7 could be achieved by obtaining a deeper Core Binding Factor (CBF) Acute plantation. 8 earlyMedicine, clearance General of blasts Hospital before “Santa clonal Maria di MyeloidMarta Campo Leukemia dell’Orto, (AML) patients Fabio Benedetti,6 Giovanni Martinelli,3 evolutionCa’ Foncello”, of the disease. piazza Ospedale Among other 1, 31100Acknowledgments: the authors are grateful to experienceFilippo Gherlinzoni treatment 1 failure in the order of attempts,Treviso, theItaly. addition of a third Leukemia-Lymphoma-Myeloma ONLUS chemotherapeutic drug, such as Fludarabine, Clinical and experimental 1 1 Association (AIL), Section of Treviso (Italy), Department of Hematology, General 2,3 40-50%. As such, efforts have been made to has Tel.:been +39.0422.322221 tested in various - Fax:clinical +39.0422.322542 trials, for the financial support. 1 1 2 efficacyMichele Gottardi, of gemtuzumabFederico Mosna, increaseHospital, the Treviso; dose Laboratory intensity of of firstCell line withE-mail: conflicting [email protected] results regarding Overall 2 Cryopreservation, General Hospital, Survival (OS), but, in most cases, with an Contributions: MG and FM designed and per- Sergio de Angeli, 3 ozogamicin in core binding treatmentTreviso; Department over of theHematology, standard increase in the rate of complete remission formed the research, analyzed data and wrote COMPARAZIONE RETROSPETTIVA DI 25 PAZIENTI CON CBF AML 2,3 Cristina Papayannidis,3 Anna Candoni,4 Daunorubicin/CytarabineInstitute “L.A. Seragnoli,” 3+7 Universityinduction, of (CR)Key and words:Disease-Free Core Survival binding (DFS). factor; acutethe manuscript; CP, AC, MC, CT, AP and FB Bologna; 4Department of Hematology, Moreover, novel agents with compatible collected data and performed research; SDA factor acute5 myeloid leukemia6 myeloid leukemia; gemtuzumab ozogamicin;performed the clonogenic assays; MCD per- Marino Clavio, Cristina Tecchio, under the assumption5 that better survival safety profile have been tested in addition to TRATTATI IN INDUZIONE CON FLAI5 University VS of 12 Udine; PAZIENTI Department CON of CBF formed the molecular tests; FM performed the 7 chemotherapy;immunotherapy; among autologous these, stem cell the trans- Andrea Piccin, couldHematology, be achieved University by obtaining of Genoa; a deeper statistical analysis; GM and FG critically 6 2 immunoconjugateplantation. Gemtuzumab AML TRATTATI CON FLAI5+MYLOTARG 3 mg/m8 earlyDepartment clearance of Hematology, of IL GIORNO 6 blasts before University clonal revised the manuscript; all authors approved Marta Campo dell’Orto, of Verona; 7Department of Hematology, Ozogamicin (GO), which combines in a the final manuscript. single agent a monoclonal antibody targeting Fabio Benedetti,6 Giovanni Martinelli,3 evolutionGeneral Hospital, of the disease.Bolzano; 8 AmongDepartment other CD33Acknowledgments: with the DNA the damagingauthors are toxingrateful toConflict of interest: the authors declare no of Pathology, General Hospital, Treviso, Filippo Gherlinzoni1 attempts, the addition of a third calicheamicin.Leukemia-Lymphoma-Myeloma Following initial studies, GO ONLUSpotential conflict of interest. chemotherapeuticItaly drug, such as Fludarabine, has been combined to chemotherapy to 1 Association (AIL), Section of Treviso (Italy), Department of Hematology, General 2,3 improve efficacy, particularly in the case of Received for publication: 4 January 2017. has been tested in various clinical trials, for the financial support. Revision received: 23 April 2017. 2 CBF AML, in which blasts highly express Hospital,CONSOLIDAMENTO CON 2-3CICLI DI ARA-C AD ALTE DOSI Treviso; Laboratory of Cell with conflicting results regarding Overall the target antigen.1,3 Three-drug Fludarabine- Accepted for publication: 27 April 2017. Cryopreservation, General Hospital, SurvivalAbstract (OS), but, in most cases, with an basedContributions: regimens combined MG and withFM designed GO proved and per-This work is licensed under a Creative successful mainly in the setting of Commons Attribution-NonCommercial 4.0 3 Leukemia-initiating cells of core binding Treviso; Department of Hematology, increase in the rate of complete remission cytogeneticallyformed the research, favorable, analyzed such data as and CBF wroteInternational License (CC BY-NC 4.0). factor (CBF) acute myeloid leukemia 2,3 AML,the or manuscript; intermediate-risk CP, AC, AML, MC, 3,4CT,whereas AP and FB Institute “L.A. Seragnoli,” University of (CR)(AML) and likely Disease-Free derive from Survival early committed (DFS). ©Copyright M. Gottardi et al., 2017 resultscollected have somehow data and been performed disappointing research; in SDA Bologna; 4Department of Hematology, Moreover,hematopoietic novel precursors agents expressing with compatible CD33. adverse risk patients.3,4 Despite this, the Licensee PAGEPress, Italy As such, targeting CD33 could ameliorate clinicalperformed development the clonogenic of GO assays; has suffered MCD per-Hematology Reports 2017; 9:7028 5 816safety profile have been tested in addition to doi:10.4081/hr.2017.7028 I PAZIENTI CON MUTAZIONE TKDUniversity of Udine; Department of the ALLA DIAGNOSI O IN CASO DI chance of cure of CBF AML patients. We fromformed concerns the molecular raised tests; by an FM increased performed the compared 12 CBF AML patients treated with Hematology, University of Genoa; chemotherapy; among these, the incidencestatistical of analysis; hepatotoxicity GM and and FG veno- critically Fludarabine, Cytarabine, Idarubicin and occlusive disease when GO was used at the mg/m2 on days1-5; Cytarabine 2 gr/m2 on PERSISTENZA 6 DEL TRASCRITTO MOLECOLARE immunoconjugate AL TERMINE Gemtuzumab DEL 2 5 Department of Hematology, University Gemtuzumab Ozogamicin (FLAI-GO doserevised of 9 mg/m the manuscript;twice during all authorsinduction, approveddays1-5; Idarubicin 10 mg/m2 on days 1,3,5; regimen) with 25 CBF AML patients treated of Verona; 7Department of Hematology, Ozogamicin (GO), which combines in a and thefrom final the manuscript. early results of a randomized GO 3 mg/m2 on day 6),4 and consolidated CONSOLIDAMENTO singlewith agent the same a monoclonal schedule, antibody but without targeting GO. phase-3 trial that showed no advantage in OS with two high-dose cytarabine (HiDAC)- General Hospital, Bolzano; 8Department With the limit of small numbers, we and a significant increase of Treatment- based cycles (overall dose 24 gr/m2/cycle). CD33observed with a consistent the DNA trend damaging toward better toxin RelatedConflict Mortality of interest: (TRM) the (5% authors vs 1%) declarein the noWe applied a regimen that was previously of Pathology, General Hospital, Treviso, 6 calicheamicin.overall survival, Following disease freeinitial survival studies, and GO GO-treatedpotential group. conflictThis of interest.ultimately led to the described in an independent series of AML Italy event free survival in the FLAI-GO group. withdrawal of the drug from the market in patients.4 Patients with c-KIT tyrosine hasWe beenalso demonstrated combined tothe chemotherapyability of GO to to 2010. Later studies3,7,8 showed, on the kinase domain mutation at codon 816 ALLOGENICO O AUTOLOGO improveinduce efficacy, the disappearance particularly in in vitro theof case the of opposite,Received an unequivocal for publication: survival 4 January benefit 2017. by (TKD816)at diagnosis (n=2) or with the Goardi M. et al Hematology Reports 2017 AML1-ETO molecular transcript in a GO,Revision even if mainly received: restricted 23 April to 2017. CBF AML, persistence of molecular transcript, assessed CBFpolymerase AML, in chain which reaction-positive blasts highly express graft at the lower schedule of 3 to 6 mg/m2, by sequential polymerase chain reaction thewithout target decreasingantigen.1,3 Three-drugthe clonogenic Fludarabine- potential withoutAccepted neither for increased publication: hepatotoxicity 27 April 2017. nor (PCR), at the end of consolidation (n=5), i.e. of CD34+/CD38- cells. This represent the higher TRM.3,7,8 This led to the revaluation Minimal Residual Disease (MRD)- Abstract basedproof regimens of principle combined for using GOwith in GO a purging proved of GO,This which, work unfortunately,is licensed under has not a Creative yet positivity, were then consolidated with either strategy before autologous stem cell resulted in the reinstitution of the drug to allogeneic or autologous hematopoietic stem successful mainly in the setting of Commons Attribution-NonCommercial5 4.0 Leukemia-initiating cells of core binding transplantation. Therefore, our data argue in clinical practice. cell transplantation (HSCT) based on the cytogeneticallyfavor of the reinstitution favorable, of such GO as in CBFthe InternationalIn order to address License the (CC role BY-NC of GO 4.0).in the availability of a donor. We decided to further factor (CBF) acute myeloid leukemia AML,therapy or intermediate-riskof CBF AML. AML,3,4 whereas treatment of CBF AML, we retrospectively intensify the treatment of KIT mutated (AML) likely derive from early committed reviewed 12 CBF AML patients [t(8;21) patients early on, based on initial studies results have somehow been disappointing in n=8;©Copyright inv(16) n=4] M. treatedGottardi from et al., 2006 2017 to 2009 showing an adverse prognosis of these hematopoietic precursors expressing CD33. adverse risk patients.3,4 Despite this, the withLicensee the FLAI-GO PAGEPress, regimen Italy (Fludarabine 30 patients as compared to KIT wild-type CBF As such, targeting CD33 could ameliorate clinical development of GO has suffered Hematology Reports 2017; 9:7028 [Hematologydoi:10.4081/hr.2017.7028 Reports 2017; 9:7028] [page 87] the chance of cure of CBF AML patients. We from concerns raised by an increased compared 12 CBF AML patients treated with incidence of hepatotoxicity and veno- Fludarabine, Cytarabine, Idarubicin and occlusive disease when GO was used at the mg/m2 on days1-5; Cytarabine 2 gr/m2 on Gemtuzumab Ozogamicin (FLAI-GO dose of 9 mg/m2 twice during induction,5 days1-5; Idarubicin 10 mg/m2 on days 1,3,5; regimen) with 25 CBF AML patients treated and from the early results of a randomized GO 3 mg/m2 on day 6),4 and consolidated with the same schedule, but without GO. phase-3 trial that showed no advantage in OS with two high-dose cytarabine (HiDAC)- With the limit of small numbers, we and a significant increase of Treatment- based cycles (overall dose 24 gr/m2/cycle). observed a consistent trend toward better Related Mortality (TRM) (5% vs 1%) in the We applied a regimen that was previously overall survival, disease free survival and GO-treated group.6 This ultimately led to the described in an independent series of AML event free survival in the FLAI-GO group. withdrawal of the drug from the market in patients.4 Patients with c-KIT tyrosine We also demonstrated the ability of GO to 2010. Later studies3,7,8 showed, on the kinase domain mutation at codon 816 induce the disappearance in vitro of the opposite, an unequivocal survival benefit by (TKD816)at diagnosis (n=2) or with the AML1-ETO molecular transcript in a GO, even if mainly restricted to CBF AML, persistence of molecular transcript, assessed polymerase chain reaction-positive graft at the lower schedule of 3 to 6 mg/m2, by sequential polymerase chain reaction without decreasing the clonogenic potential without neither increased hepatotoxicity nor (PCR), at the end of consolidation (n=5), i.e. of CD34+/CD38- cells. This represent the higher TRM.3,7,8 This led to the revaluation Minimal Residual Disease (MRD)- proof of principle for using GO in a purging of GO, which, unfortunately, has not yet positivity, were then consolidated with either strategy before autologous stem cell resulted in the reinstitution of the drug to allogeneic or autologous hematopoietic stem transplantation. Therefore, our data argue in clinical practice.5 cell transplantation (HSCT) based on the favor of the reinstitution of GO in the In order to address the role of GO in the availability of a donor. We decided to further therapy of CBF AML. treatment of CBF AML, we retrospectively intensify the treatment of KIT mutated reviewed 12 CBF AML patients [t(8;21) patients early on, based on initial studies n=8; inv(16) n=4] treated from 2006 to 2009 showing an adverse prognosis of these with the FLAI-GO regimen (Fludarabine 30 patients as compared to KIT wild-type CBF

[Hematology Reports 2017; 9:7028] [page 87] AML.9 Conversely, the persistence of the performed in 2 patients because of the lack numbers. The achievement of MRD- Briefmolecular Report transcript at the end of of cytogenetic response after induction negativity (FLAI-GO=100% of the 4 consolidation was considered a predictor of therapy. patients analyzed; FLAI=63% of the 13 adverse prognosis based on previous Patients in the two groups reached a patient analyzed), assessed by PCR, was of experiences,10,11 as well as our own comparable rate of CR after induction pivotal prognostic importance (P<0.001 for unpublished data. (n=12/12 vs 22/25, P=0.540); no death in either OS, DFS and EFS). We compared this group with 25 CBF induction was observed, and all patients The potential of GO in the treatment of AML patients [t(8;21) n=13; inv(16) n=12] completed their therapeutic schedule. At CBF AML is based on strong biological treated according to the same criteria and median follow-up of 69.2 months, 3 patients bases: first of all, most t(8;21) AML blasts with the same schedule, but without GO, in in the FLAI-GO group relapsed at 8, 14 and do not express the transporter P-glycoprotein the years 2003-2006 and 2010-2013. The 42 months after the achievement of CR; of (Pgp),12 i.e. the Multi-Drug Resistance 1 two groups were comparable in all clinical these, two out of three achieved second CR (MDR1) gene product, and this seems and laboratory features (Table 1). In the following rescue therapy and underwent related to a selective repression of the latter group, autologous HSCT was allogeneic HSCT. Conversely, at median promoter of MDR1 by AML1-ETO.13 performed in 5 patients because of MRD- DFS 14.7 months, 11 patients of the FLAI Several studies have pointed out how GO positivity, and allogeneic HSCT was group relapsed; among these, 7 out of 10 extrusion by Pgp affects clinical response.14 achieved second CR and were consolidated Moreover, CBF fusion transcripts promote with allogeneic HSCT. leukemogenesis by inducing the initial As shown in Figure 1, we observed a expansion of a preleukemic myeloid cell consistent, yet non-statistical trend towards compartment predisposed to secondary better OS, DFS and, most importantly in mutations and characterized by CD33+ early evaluating the efficacy of first line therapy, committed myeloid precursors incorporating Event Free Survival (EFS) in the FLAI-GO the AML1-ETO or CBFB-MYH11 group (5-yrs OS 69.4% vs 48.6%, P=0.202; transcripts.15 According to this model, 5-yrs DFS 54.7% vs 42.4%, P=0.327; 5-yrs founding Leukemia-Initiating Cells of CBF EFS 54.7% vs 36.9%, P=0.136; Figure 1). AML, differently from other types of AML, Besides, patients tended to relapse later would arise from these early committed when treated with FLAI-GO (median DFS: myeloid precursors rather than the more unreached vs 14.7 months; Figure 1). We immature Hematopoietic Stem Cells,15 and believe these differences did not reach the therefore be more sensitive to GO therapy statistical significance mainly due to small because of their markedly higher expression

Table 1. Baseline patient characteristics. FLAI5 My-FLAI5 P Median age 41.3 (18-66) 46.3 (29-67) 0.2602 Sex 12 M + 13 F 6 M + 6 F 0.909 Secondary acute myeloid leukemia 0 0 NA Hepatomegaly 4 2 1.0 Splenomegaly 3 2 1.0 Sarcoma 0 1 1.0 Hemoglobin gr/dL 8.4 (4.2-11) 8.5 (5-13.6) 0.9100 White blood cells ×103/L 19.0 (1.6-95) 18.7 (4.5-45.5) 0.9706 N ×103/L 1.86 (0.33-6.35) 2.58 (0.45-11.36) 0.3680 Mo ×103/L 0.95 (0.01-2.56) 0.58 (0.01-2.68) 0.1330 Figure 1. Survival by treatment with Gemtuzumab Ozogamicin (GO). Patients Ly ×103/L 2.67 (0.50-6.80) 2.86 (0.60-7.73) 0.8016 receiving FLAI-GO induction therapy Blasts ×103/L 10.9 (0.01-65.55) 11.6 (0.22-32.0) 0.9074 showed a consistent, yet nonstatistical, Platelets ×103/L 60.66 (8-255) 73.72 (6-531) 0.7150 trend towards better Overall Survival (OS), Disease-Free Survival (DFS) and, most Elevated LDH 18 10 0.638 importantly in determining the effect of DIC 2 2 0.305 first line treatment, Event-Free Survival (EFS), as compared to a group of patients Acute renal failure 0 1 0.314 treated with FLAI. Numbers of the two t(8;21)/inv(16) 13/12 8/4 0.491 groups are showed as n; only patients FLT3-ITD 3 2 1.0 achieving complete remission after induc- tion therapy were considered in determin- NPM1 mutated 0 0 NA ing DFS. Inclusion criteria and treatment KIT TKD 816 mutated 1 1 1.0 schedule was the same in the two groups Packed BM (>80%) 12 5 0.717 apart from the addition of GO at 3 mg/m2 to induction therapy at day 6 of the FLAI- Additional cytogenetic abnormalities None: 14 pts; 1: 7 pts; None: 4; 1: 5 pts; 0.387 GO regimen. 2: 3 pts; 3: 1 pts 2: 1 pts; 3: 2 pts

[page 88] [Hematology Reports 2017; 9:7028] Goardi M. et al Hematology Reports 2017 RISULTATI

FLAI5 My-FLAI5

RC DOPO INDUZIONE 22/25 (88%) 12/12 (100%) RICADUTE 11/22 (50%) 3/12 (25%) Follow up 69.2 mesi

• NESSUNA MORTE TOSSICA

• TUTTI I PAZIENTI HANNO COMPLETATO LA PREVISTA SCHEDULA DI TRATTAMENTO

Goardi M. et al Hematology Reports 2017 AML.9 Conversely, the persistence of the performed in 2 patients because of the lack numbers. The achievement of MRD- Briefmolecular Report transcript at the end of of cytogenetic response after induction negativity (FLAI-GO=100% of the 4 consolidation was considered a predictor of therapy. patients analyzed; FLAI=63% of the 13 adverse prognosis based on previous Patients in the two groups reached a patient analyzed), assessed by PCR, was of experiences,10,11 as well as our own comparable rate of CR after induction pivotal prognostic importance (P<0.001 for unpublished data. (n=12/12 vs 22/25, P=0.540); no death in either OS, DFS and EFS). We compared this group with 25 CBF induction was observed, and all patients The potential of GO in the treatment of AML patients [t(8;21) n=13; inv(16) n=12] completed their therapeutic schedule. At CBF AML is based on strong biological treated according to the same criteria and median follow-up of 69.2 months, 3 patients bases: first of all, most t(8;21) AML blasts with the same schedule, but without GO, in in the FLAI-GO group relapsed at 8, 14 and do not express the transporter P-glycoprotein the years 2003-2006 and 2010-2013. The 42 months after the achievement of CR; of (Pgp),12 i.e. the Multi-Drug Resistance 1 two groups were comparable in all clinical these, two out of three achieved second CR (MDR1) gene product, and this seems and laboratory features (Table 1). In the following rescue therapy and underwent related to a selective repression of the latter group, autologous HSCT was allogeneic HSCT. Conversely, at median promoter of MDR1 by AML1-ETO.13 performed in 5 patients because of MRD- DFS 14.7 months, 11 patients of the FLAI Several studies have pointed out how GO positivity, and allogeneic HSCT was group relapsed; among these, 7 out of 10 extrusion by Pgp affects clinical response.14 achieved second CR and were consolidated Moreover, CBF fusion transcripts promote with allogeneic HSCT. leukemogenesis by inducing the initial As shown in Figure 1, we observed a expansion of a preleukemic myeloid cell consistent, yet non-statistical trend towards compartment predisposed to secondary better OS, DFS and, most importantly in mutations and characterized by CD33+ early evaluating the efficacy of first line therapy, committed myeloid precursors incorporating Event Free Survival (EFS) in the FLAI-GO the AML1-ETO or CBFB-MYH11 group (5-yrs OS 69.4% vs 48.6%, P=0.202; transcripts.15 According to this model, 5-yrs DFS 54.7% vs 42.4%, P=0.327; 5-yrs founding Leukemia-Initiating Cells of CBF EFS 54.7% vs 36.9%, P=0.136; Figure 1). AML, differently from other types of AML, Besides, patients tended to relapse later would arise from these early committed when treated with FLAI-GO (median DFS: myeloid precursors rather than the more unreached vs 14.7 months; Figure 1). We immature Hematopoietic Stem Cells,15 and believe these differences did not reach the therefore be more sensitive to GO therapy statistical significance mainly due to small because of their markedly higher expression

Table 1. Baseline patient characteristics. FLAI5 My-FLAI5 P Median age 41.3 (18-66) 46.3 (29-67) 0.2602 Sex 12 M + 13 F 6 M + 6 F 0.909 Secondary acute myeloid leukemia 0 0 NA Hepatomegaly 4 2 1.0 Splenomegaly 3 2 1.0 Sarcoma 0 1 1.0 Hemoglobin gr/dL 8.4 (4.2-11) 8.5 (5-13.6) 0.9100 White blood cells ×103/L 19.0 (1.6-95) 18.7 (4.5-45.5) 0.9706 N ×103/L 1.86 (0.33-6.35) 2.58 (0.45-11.36) 0.3680 Mo ×103/L 0.95 (0.01-2.56) 0.58 (0.01-2.68) 0.1330 Figure 1. Survival Goardi M. et al Hematology Reports 2017 by treatment with Gemtuzumab Ozogamicin (GO). Patients Ly ×103/L 2.67 (0.50-6.80) 2.86 (0.60-7.73) 0.8016 receiving FLAI-GO induction therapy Blasts ×103/L 10.9 (0.01-65.55) 11.6 (0.22-32.0) 0.9074 showed a consistent, yet nonstatistical, Platelets ×103/L 60.66 (8-255) 73.72 (6-531) 0.7150 trend towards better Overall Survival (OS), Disease-Free Survival (DFS) and, most Elevated LDH 18 10 0.638 importantly in determining the effect of DIC 2 2 0.305 first line treatment, Event-Free Survival (EFS), as compared to a group of patients Acute renal failure 0 1 0.314 treated with FLAI. Numbers of the two t(8;21)/inv(16) 13/12 8/4 0.491 groups are showed as n; only patients FLT3-ITD 3 2 1.0 achieving complete remission after induc- tion therapy were considered in determin- NPM1 mutated 0 0 NA ing DFS. Inclusion criteria and treatment KIT TKD 816 mutated 1 1 1.0 schedule was the same in the two groups Packed BM (>80%) 12 5 0.717 apart from the addition of GO at 3 mg/m2 to induction therapy at day 6 of the FLAI- Additional cytogenetic abnormalities None: 14 pts; 1: 7 pts; None: 4; 1: 5 pts; 0.387 GO regimen. 2: 3 pts; 3: 1 pts 2: 1 pts; 3: 2 pts

[page 88] [Hematology Reports 2017; 9:7028] PBSC PURGING WITHAutologous StemMYLOTARG? Cell Transplantation with PCR-Negative Graft Would Be Associated with Autologous Stem Cell Transplantation with PCR-Negativea Favorable OutcomeGraft Would in Core-Binding Be Associated Factor with a FavorableAcute Outcome Myeloid in Core-Binding Leukemia Factor Hideki Nakasone,Acute1 Koji Myeloid Izutsu,1 Satoshi Leukemia Wakita,2 Hiroki Yamaguchi,2 1 1 Hideki Nakasone,Michiko1 Koji Muramatsu-Kida, Izutsu,1 SatoshiKensuke Wakita,2 UsukiHiroki Yamaguchi,2 Michiko Muramatsu-Kida,1 Kensuke Usuki1 Although core-binding factor acute myeloid leukemia (CBF-AML) is generally considered to be a low-risk Although core-binding factor acute myeloid leukemia (CBF-AML) is generally considered to be a low-risk form ofform AML, of AML, the survival the survival rate rate is still is still 50% 50% to to 60%. 60%. Toevaluate Toevaluate the the effectiveness effectiveness of ofautologous autologous stem stem cell trans- cell trans- plantationplantation (ASCT) (ASCT) with with a PCR-negative a PCR-negative graft graft we we analyzed analyzed a series series of of consecutive consecutive CBF-AML CBF-AML patients. patients. Between Between 19971997 and 2006, and 2006, 18 patients 18 patients aged aged\\6060 years years were were referred referred under under a a diagnosis diagnosis of of CBF-AML. CBF-AML. Peripheral Peripheral blood blood stem cells (PBSC) were collected after a second or further course of postremission therapy. When .2.0 6 stem cells (PBSC) were collected after a second or further course of postremission therapy. When .2.0  10 /kg CD34-positive6 cells with minimal residual disease (MRD) undetectable by nested polymerase chain reaction10 /kg (PCR) CD34-positive had been collected, cells with minimalASCTwas residual performed disease with (MRD) busulfan, undetectable etoposide, by nested and polymerase cytarabine combinedchain with granulocytereaction (PCR) colony-stimulating had been collected, factor. ASCTwas Event-free performed survival with busulfan, (EFS) and etoposide, complications and cytarabine of ASCT combined were then assessed.with Fourteen granulocyte of colony-stimulating the 18 patients factor. received Event-free ASCT. survival The median (EFS) and observation complications period of ASCT was were4.4 years. then The 5-yearassessed. EFS was Fourteen 93% for of ASCT the 18 patients, patients received despite ASCT. the presence The median of adverse observation factors. period In was 8 of 4.4 10 years. patients The who had detectable MRD in the bone marrow before ASCT, MRD became undetectable after ASCT. Neutrophils recovered5-year promptly EFS was 93% within for 2ASCT weeks, patients, but platelets despite the recovered presence relatively of adverse slowly. factors. Half In 8 of of the 10 patients patients who suffered fromhad varicella detectable zoster MRD virus in the infection. bone marrow Although before 1 ASCT, case MRD of myelodysplastic became undetectable syndrome after ASCT. occurred, Neutrophils there was no caserecovered of relapse. promptly ASCT within with 2 a weeks, PCR-negative but platelets graft recovered was associated relatively with slowly. excellent Half of the EFS. patients For patients suffered with CBF-AML,from varicella especially zoster with virus adverse infection. factors Although or remnant 1 case MRD of myelodysplastic in the bone marrow, syndrome this occurred, strategy there is the was treat- mentno of choice.case of relapse. ASCT with a PCR-negative graft was associated with excellent EFS. For patients with Biol BloodCBF-AML, Marrow especially Transplant with 14: adverse 1262-1269 factors (2008) or remnantÓ 2008 MRD American in the Society bone marrow, for Blood this and strategy Marrow is Transplantation the treat- KEY WORDS:ment of choice.Core binding factor acute myeloid leukemia, Autologous stem cell transplantation, Minimal residual disease, Polymerase chain reaction Biol Blood Marrow Transplant 14: 1262-1269 (2008) Ó 2008 American Society for Blood and Marrow Transplantation Nakasone H. et al Biol Blood Marrow Transplant 2008 KEY WORDS: Core binding factor acute myeloid leukemia, Autologous stem cell transplantation, Minimal INTRODUCTION residual disease, Polymerase chain reaction fuses the AML1 (CBFa2) gene located on chromo- some 21 to the ETO (MTG8) gene located on chro- Translocation (8;21) (q22;q22) or inversion (16) mosome 8. The CBFb gene located at 16q22 fuses occurs in approximately 7% to 8% of patients with with the MYH11 gene located at 16p13. The AML1- de novo acute myeloid leukemia (AML) [1]. These ETO or CBFb-MYH11 fusion protein represses and leukemiaINTRODUCTION entities are associated with aberration of fusesalters the the AML1 function (CBF ofa CBF2) gene during located normal on chromo- differentia- core-binding factors (CBF), which are heterodimeric sometion [2] 21. to the ETO (MTG8) gene located on chro- transcriptionalTranslocation regulators (8;21) (q22;q22)containing or inversion a common (16) mosomeBoth 8. cytogenetic The CBFb gene groups located (referred at 16q22 to fuses as CBF- b (CBFb) and 1 of 3 a (CBFa) subunits. Translocation occurs in approximately 7% to 8% of patients with withAML) the have MYH11 a relatively gene located favorable at 16p13. prognosis The AML1- compared with most other forms of adult AML [1,3-5]. In youn- de novo acute myeloid leukemia (AML) [1]. These ETO or CBFb-MYH11 fusion protein represses and leukemia entities are associated with aberration of ger patients, repeated cycles of high-dose cytarabine alters(HDAC) the function therapy of can CBF prolong during survival normal differentia-[6,7]. From thecore-binding1Division of factors Hematology, (CBF), Kanto which Medical are heterodimeric Center NTT EC, Tokyo, Japan; and 2Division of Hematology, Department tion Prognostic[2]. factors of CBF-AML have been evalu- of Internaltranscriptional Medicine, regulators Nippon Medical containing School, Tokyo, a common Japan. atedBoth in several cytogenetic studies. groups In t(8;21) (referred AML to patients, as CBF- infe- Correspondenceb (CBFb) and and reprint 1 of 3 a requests:(CBFa) Hidekisubunits. Nakasone, Translocation M.D., rior outcome has been associated with a high white Division of Hematology, Kanto Medical Center NTT EC, 5- AML)blood have cell a (WBC) relatively count favorable[8] prognosis, a low comparedplatelet count 9-22 Higashigotanda, Shinagawa-ku, Tokyo, Japan 141-8625 with[8,9] most, a high other WBC forms index of adult[10] AML, loss[1,3-5] of sex. chromosomes In youn- (e-mail: [email protected]). Received July 3, 2008; accepted August 25, 2008 ger[8], patients, expression repeated of CD56 cycles antigen of high-dose[11], extramedullarycytarabine 1083-8791/08/1411-0001$34.00/0 disease [12], non-White race [9], and older age [9]. 1 (HDAC) therapy can prolong survival [6,7]. doi:10.1016/j.bbmt.2008.08.012From the Division of Hematology, Kanto Medical Center NTT In inv(16) AML patients, a high WBC count [13,14], EC, Tokyo, Japan; and 2Division of Hematology, Department Prognostic factors of CBF-AML have been evalu- 1262 of Internal Medicine, Nippon Medical School, Tokyo, Japan. ated in several studies. In t(8;21) AML patients, infe- Correspondence and reprint requests: Hideki Nakasone, M.D., rior outcome has been associated with a high white Division of Hematology, Kanto Medical Center NTT EC, 5- blood cell (WBC) count [8], a low platelet count 9-22 Higashigotanda, Shinagawa-ku, Tokyo, Japan 141-8625 [8,9], a high WBC index [10], loss of sex chromosomes (e-mail: [email protected]). Received July 3, 2008; accepted August 25, 2008 [8], expression of CD56 antigen [11], extramedullary 1083-8791/08/1411-0001$34.00/0 disease [12], non-White race [9], and older age [9]. doi:10.1016/j.bbmt.2008.08.012 In inv(16) AML patients, a high WBC count [13,14],

1262 of CD33. The distinctive chemosensitivity been related to a better prognosis by many proved able to induce the near-complete shown by CBF AML1 could also be trials. 19 Autologous HSCT, performed at the saturation of the CD33 antigen Brief sites. Report20 Cells explained by these biological differences.15 end of first line treatment, might be were then collected, GO removed by As such, autologous HSCT could be beneficial for selected CBF AML patients to centrifugation, and cells reseeded in Petri used to increase the dose intensity of first achieve the disappearance of MRD, dishes containing semisolid MethoCult GF- line therapy in selected patients, improving especially when performed using MRD- H4434 medium. The number of OS, as shown by some studies.16 Results are negative Peripheral Blood Stem Cells Colony-Forming Units (CFU) was best when MRD-negativity in the bone (PBSC).17 Monoclonal antibodies are thus determined after 14 days incubation. marrow and in the products of leukapheresis being tested to provide in vivo purging of Using unsorted cells, we observed a is obtained before transplantation.17 PBSC. This approach resembles the use of significant decrease in the number of CFU- Recently, a 5-year EFS of 93% was reported Rituximab in the therapy of patients affected GEMM (clonal efficiency 0.000697 vs in a small series of CBF AML patients by CD20+ lymphomas and undergoing 0.01037, P=0.016), CFU-GM (0.002606 vs undergoing autologous HSCT with a PCR- autologous HSCT. Concerns are there, 0.004071, P=0.038) and BFU-E colonies negative graft, with the disappearance of though, that treatment with GO might affect (0.003213 vs 0.004697, P=0.031) in the cells CBF transcripts after HSCT in 8 out of 10 hematopoietic reconstitution by reducing the exposed to GO. We then performed the same previously MRD-positive patients.18 number of hematopoietic long-term experiments on samples from the same Nonetheless, in a previous series we repopulating cells collected by PBSC units after the enrichment in observed that clinical outcome of patients leukapheresis. CD34+/CD38- cells by immunomagnetic could be improved also when autologous We therefore evaluated the clonogenic sorting. The efficacy of this sorting was HSCT had been performed with CD34+ growth of PBSC collected at the end of proved by either immunophenotypic analysis 16 grafts with persistent molecular transcripts. consolidation therapy from five patients with and functional assays, which resulted in We explain this finding by the higher overall CBF AML. Briefly, mononuclear cells were significantly more numerous CFU-GEMM, dose-intensity achieved by first line cultured in RPMI medium in the presence or CFU-GM, BFU-E (data not shown). This treatment by including autologous HSCT. absence of GO at a concentration of 5 µg/mL time, though, we did not observe a The achievement of MRD-negativity has for two hours, a previously in vitro dose significant decrease in the clonogenic potential of CD34+/CD38- cells by the exposure to GO (0.01518 vs 0.02631, P=0.351), thus confirming the preservation of more immature hematopoietic precursors. Moreover, in one MRD-positive patient we could observe the disappearance of the AML1-ETO molecular transcript following in vitro incubation with GO at a concentration of 5 µg/mL for two hours (Figure 2). Therefore, with the limit of small numbers, the results that we report suggest the possibility of using GO in a purging strategy that would possibly act on residual CBF AML cells without affecting the repopulating ability of PBSC. In order to avoid the limitation of in vitro purging, GO could be used in vivo before CD34+ cell collection in MRD-positive patients.

References 1. Paschka P. Core binding factor acute myeloid leukemia. Semin Oncol 2008; 35:410-7. 2. Borthakur G, Kantarjian H, Wang X, et al. Treatment of core-binding-factor acute myelogenous leukemia with fludarabine, cytarabine and granulocyte Figure 2. Disappearance of AML1-ETO molecular transcript following in vitro purging colony-stimulating factor results in GO PUO’ FUNZIONARE of PBSC with Gemtuzumab COME Ozogamicin. PURGING Samples from PBSC collected IN from VITRO 5 patients (O improved ANCHE event-free IN survival. Cancer affected by CBF AML at the end of consolidation were cultured in the presence or 2008;113:3181-5. absence of Gemtuzumab Ozogamicin at a concentration of 5 µg/mL for two hours. In one patient (i.e. TG) affected by t(8;21) AML with persistent AML1-ETO transcript at 3. Burnett AK, Hills RK, Milligan D, et al. VIVO PRIMA the DELLA end of consolidation RACCOLTA and in the PBSC, incubation DELLE with GO PBSC obtained the NEI disappear- PAZIENTI Identification MRD- of patients with acute ance of cells expressing AML1-ETO, as tested by either direct and nested PCR. Abelson myeloblastic leukemia who benefit from POSITIVI), SENZA (ABL) amplification MENOMARE was used as internal control. IL TG: initials POTENZIALE of the patient’s name; Ctrl- CLONOGENICO the addition of gemtuzumab /Ctrl+: negative and positive controls. ozogamicin: results of the MRC AML 15 DELLE CELLULE CD34+/CD38- [Hematology Reports 2017; 9:7028] [page 89] Goardi M. et al Hematology Reports 2017 CONCLUSIONI

• DOPO CIRCA 25 ANNI DI UN PERCORSO TRAVAGLIATO DI STUDI PRE-CLINICI E CLINICI, MYLOTARG HA TROVATO LA SUA APPROVAZIONE DEFINITIVA NELLE AML CD33+, ALLA DIAGNOSI O CON MALATTIA RICADUTA/REFRATTARIA, NEI PAZIENTI PEDIATRICI, ADULTI O ANZIANI, IN COMBINAZIONE O SINGLE-AGENT

• IL PROFILO DI TOSSICITA’ E’ ASSOLUTAMENTE ACCETTABILE CON DOSI RIDOTTE E FRAZIONATE

• EFFICACE SOPRATTUTTO NEI PAZIENTI A RISCHIO BASSO O INTERMEDIO, E PARTICOLARMENTE NELLE CBF-LEUKEMIAS

• MECCANISMI DI RESISTENZA • PURGING IN VITRO O IN VIVO? SVILUPPI FUTURI • NUOVI AGENTI CHE ABBIANO COME TARGET CD33

VADASTUXIMAB TALARINE (SGN-33A)

ANTICORPO LINTUZUMAB CONIUGATO A 2 MOLECOLE DI UN DIMERO PIRROLOBENZODIAZEPINICO RISULTATI MOLTO PROMETTENTI IN ASSOCIAZIONE AD IPOMETILANTI, MA ELEVATA TOSSICITA’

RADIOIMMUNOTERAPIA CON 225Ac-LINTUZUMAB

• NUOVI AGENTI CHE ABBIANO COME TARGET ALTRE MOLECOLE DIVERSE DAL CD33 CD123

• SL-401 = ANTICORPO+TOSSINA DIFTERICA