Brown 2014

Novel Treatments for Chronic Lymphocytic Leukemia andMoving ForwardJennifer R. Brown, MD, PhD, David L. Porter, MD, and Susan M. OBrien, MD

OVERVIEW

The last several years have seen an explosion of novel therapies for chronic lymphocytic leukemia (CLL). These include the antibodyobintutuzumab (GA-101), as well as small-molecule inhibitors of key pathways involved in the pathogenesis of CLL, specically the B-cellreceptor (BCR) pathway (especially Brutons tyrosine kinase [BTK] and P13K), and the antiapoptotic pathway (especially BCL-2). We willconsider each in turn, focusing on the molecules most advanced in clinical development. There has also been extensive developmentin rewiring the patients own immune system to treat CLL. This has been done through modifying autologous T cells to express achimeric antigen receptor (CAR). Thus far all CAR-T preparations have targeted the CD19 antigen. This is a good rational for B-cellmalignancies as CD19 expression is limited to B-cell malignancies and normal B cells. The in vivo amplication of the transduced T cellsrelies on signaling and co-signaling domains and provides signicant killing of CLL cells. As exciting as these novel agents andapproaches are, they obviously beg the question, will chemotherapy as a treatment for CLL soon be obsolete? Although chemotherapyis associated with known short-term toxicities, it has the advantage of being completed in a short period of time and being relativelyinexpensive in comparison to novel therapies. In addition, long-term follow-up of results with chemoimmunotherapy have now identieda group of patients whose remissions are maintained for more than 10 years. An important question that will arise going forward ishow to incorporate novel agents without eliminating the long term benets possible with chemoimmunotherapy in a subset of patientswith CLL.

The last several years have seen an explosion of novel ther-apies for chronic lymphocytic leukemia (CLL). Theseinclude the antibody obinutuzumab (GA-101), as well assmall-molecule inhibitors of key pathways involved in thepathogenesis of CLL, specifcally the B-cell receptor (BCR)pathway (especially Brutons tyrosine kinase [BTK] andPI3K), and the antiapoptotic pathway (especially BCL-2).Wewill consider each in turn, focusing on the molecules mostadvanced in clinical development.

ANTI-CD20 ANTIBODY: OBINUTUZUMABObinutuzumab (GA101) is a humanized, glycoengineeredtype II IgG1 antibody against CD20 which has a differentmechanism of action compared to rituximab or ofatu-mumab. Obinutuzumab causesmore direct cell killing, likelythrough a lysosomal mediated mechanism, and confers en-hanced antibody-dependent cellular cytotoxicity (ADCC),with reduced complement dependent cytotoxicity (CDC).1Obinutuzumab has shown enhanced effcacy in killing CLLcells in vitro compared to rituximab and ofatumumab. Earlyclinical data with obinutuzumab demonstrated very rapid

clearance of B cells from peripheral blood, as well as neutro-penia.2The defnitive registration trial for obinutuzumab was un-

dertaken by the German CLL Study Group in their CLL11study.3 This trial was performed in 781 previously untreatedpatients with CLL and clinically signifcant medical comor-bidities and/or reduced renal function. The patients wererandomly assigned 1:2:2 to chlorambucil (clb) alone,chlorambucil with rituximab (R-clb), or chlorambucil withobinutuzumab (G-clb). Both R-clb and G-clb treatment re-sulted in statistically signifcant improvements in overall re-sponse rate (ORR) and progression-free survival (PFS)compared to clb alone (PFS R-clb 16.3 m vs. clb 11.1 m, HR0.44, p 0.001; G-clb 26.7 m vs. clb 11.1 m, HR 0.18, p 0.001), aswell as in complete remissions (CR). Based on thesedata, G-clb was approved by theU.S. Food andDrugAdmin-istration (FDA) in November 2013 for the initial therapy forpatients with CLL. Updated data on the G-clb arm in com-parison to clb alone have also demonstrated an overall sur-vival (OS) beneft favoring G-clb (9% deaths for G-clb vs.20% deaths for clb; hazard ratio [HR] 0.41; p 0.002).Likely of greater interest, however, was the direct compar-

From the Chronic Lymphocytic Leukemia Center, Dana-Farber Cancer Institute, Boston, MA; Blood and Marrow Transplantation, Hospital of the University of Pennsylvania, Philadelphia, PA;Department of Leukemia, Division of Cancer Medicine, University of Texas MD Anderson Cancer Center, Houston, TX.

Disclosures of potential conicts of interest are found at the end of this article.

Corresponding author: Susan M. OBrien, MD, Department of Leukemia, Division of Cancer Medicine, University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 428, Houston, TX77030; email: [email protected].

2014 by American Society of Clinical Oncology.

NOVEL TREATMENTS FOR CHRONIC LYMPHOCYTIC LEUKEMIA

asco.org/edbook | 2014 ASCO EDUCATIONAL BOOK e317

ison of G-clb to R-clb. The safety of obinutuzumab waslargely similar to rituximab, with the exception of an in-creased rate of infusion reactions during the frst dose. Neu-tropenia was similar (33% with G-clb vs. 28% with R-clb).Infusion reactions may be minimized by dividing the frstdose (100 mg on day 1, 900 mg on day 2), and by premedi-cating with methylprednisolone or dexamethasone, diphen-hydramine, and acetaminophen. This comparison showedthat G-clb improvedORR, CR, andmolecular response com-pared to R-clb, and resulted in a marked increase in PFS,from16.3monthswithR-clb to 26.7monthswithG-clb.3 Thelarge difference inPFS between these anti-CD20 antibodies isimpressive, and suggests that the difference in molecularmechanism between obinutuzumab and rituximab is likelysignifcant in CLL. These results provide the impetus for fur-ther studies of obinutuzumab inCLL in relapsed disease or incombination with other novel agents.

B-CELL RECEPTOR PATHWAY INHIBITIONAlthough activating somatic mutations in the BCR pathwayare relatively rare inCLL,4 constitutive activation of the path-way is quite common, and kinases within this pathway haveproven to be excellent therapeutic targets in CLL. BTK andPI3K have been most successfully targeted in the clinic andwill therefore be the focus of this discussion, but inhibitors ofSYK andLYNmay follow soon. All inhibitors of this pathwayin the clinic result in an initial redistribution of lymphocytesinto peripheral blood which resolves slowly over time, suchthat response criteria have been modifed to include a cate-gory PR-L, meaning partial response with ongoing lympho-cytosis.5

BTK INHIBITIONBTK is a Tec family kinase that functions in signal transduc-tion downstream of the B-cell receptor. Its critical impor-tance is underscored by the fact that mutations of BTK resultin the human disease X-linked agammaglobulinemia, whichis characterized by immunodefciency and an absence of Bcells. The BTK inhibitor ibrutinib has recently shown re-markable effcacy in CLL, and two additional BTK inhibitorshave also now reported clinical data in CLL (Table 1).

IbrutinibIbrutinib is a potent covalent inhibitor that binds irreversiblyto cysteine 481 (Cys481) in BTK. Despite a relatively shorthalf-life, this irreversible inhibition results in 24-hour targetinhibition, as demonstrated in the phase I study of ibrutinibin B-cell malignancies.6 Ibrutinib is not a specifc inhibitorbut rather inhibits 19 other kinases at inhibitory concentra-tions (IC50s) less than 100 nmol/L, including six that sharea cognate cysteine and could potentially also be inhibitedcovalently.7 In fact, recent research has demonstrated exper-imentally that ibrutinib does inhibit interleukin-2 induciblekinase (ITK), a critical kinase involved in T cell signaling,covalently.8 The relative signifcance of different targets tothe clinical activity of ibrutinib is unknown, although resis-tancemutations have been found inwhichCys481 ismutatedto Ser, suggesting that BTK is at least a key target.9In the initial phase I study in B-cell malignancies, promis-

ing effcacy signals were seen in multiple diseases includingCLL,6 leading to initiation of the phase Ib/II program. Thisphase Ib/II program included cohorts of patients with re-lapsed refractory CLL treated at 420 mg (51 patients) or 840mg daily (34 patients) as well as a previously untreated oldercohort treated primarily at 420 mg daily (31 patients); theresults of both have been recently published.10,11 Interna-tional Workshop on CLL (IWCLL) response rates were 71%in both cohorts, with an additional 13% in each study havingPR-L. Most impressively, the PFS in the relapsed refractorycohort is 75% at 26 months,10 and 96% in the untreated co-hort at 22 months.11 The relapse rate is higher in those pa-tients with 17p deletion however, with 57% PFS at 26months.10 Ibrutinib was well-tolerated with the primary po-tential safety concern being bleeding risk, particularly in thesetting of anticoagulation.The RESONATE registration trial comparing ibrutinib to

ofatumumab in patients with relapsedCLLwas stopped earlyin January 2014 as it met its primary PFS endpoint as well asits secondary OS endpoint. Ibrutinib will likely be approvedby the FDA for relapsed CLL in early 2014. Combinationstudies of ibrutinib with rituximab, with bendamustine/rituximab (BR) and with fludarabine/cyclophosphamide/rituximab (FCR) have all been reported in abstract form.Additional registration trials, with ibrutinib given as a singleagent in 17p-deleted CLL and for up-front therapy in elderlypatients, as well as in combination with BR, are ongoing.

KEY POINTS

B-cell receptor inhibitors, such as ibrutinib and idelalisib,result in rapid reduction in lymphadenopathy with asimultaneous rise in the absolute lymphocyte count whichshould not be mistaken for progressive disease.

ABT-199, a novel BCL-2 inhibitor, results in marked efcacyin patients with relapsed and refractory disease but alsocan result in tumor lysis.

All trials with targeted oral agents have thus far beendesigned with continuous administration compared with theshort specied number of cycles seen with chemotherapy.

Chimeric antigen receptor (CAR)modied T-cell therapy isassociated with prolonged remissions in patients withrefractory disease; toxicities include a cytokine releasesyndrome that may be abrogated by antiinterleukin-6 (IL-6) monoclonal antibodies.

Fludarabine/cyclophosphamide/rituximab (FCR) can result inremissions lasting 10 years in a subset of patients withmutated IgVH genes and absence of poor cytogenetics suchas 17p deletion or 11q deletion.

BROWN, PORTER, AND OBRIEN

e318 2014 ASCO EDUCATIONAL BOOK | asco.org/edbook

OTHER BTK INHIBITORSTwo other covalent inhibitors of BTK that bind to the sameCys481 residue have reported clinical data. CC-292 is a morespecifc inhibitor of BTK that, unlike ibrutinib, does not in-hibit SRC family kinases or ITK. The phase I study of CC-292in CLL was recently updated12 and the drug has been welltolerated. The recommended phase II dose is 500 mg twicedaily. At this dose the IWCLL partial remission (PR) rate was50%, with 13% additional PR-L. Median follow-up is stillshort, currently at 7.4 months for this cohort as of December2013. Combination studies have been initiated. The othercompound is ONO-4059, also a more specifc inhibitor thanibrutinib. This drug is early in its phase I study, with 19 pa-tients enrolled and no dose-limiting toxicities yet observed.13IWCLL response rate was 78%, with an additional 11%PR-L.Longer follow-up will be required to clarify how the activityof either of these drugs compares to that of ibrutinib.

PI3K INHIBITORSIdelalisibThe frst PI3K inhibitor to be tested in CLL was idelalisib(formerly known as CAL-101 or GS-1101). Idelalisib is a po-tent specifc inhibitor of the delta isoform of PI3 kinase,which appears to have its most critical function in B cells.Idelalisib was studied in a large phase I study in hematologicmalignancies which included 54 patients with CLL who hadbeen heavily pretreated, with amedian of fve prior regimens,and 70%of whomhad refractory disease. The nodal responserate was 81%, IWCLL PR rate was 39%, and PR-L rate was33%. The PFS in the entire study population was 17.1

months, but for those treated at the recommended phase IIdose of 150 mg twice daily or higher, the median PFS was 29months.14 A phase II study of idelalisib with rituximab in 64previously untreated CLL patients showed a 97% ORR with19% complete remissions and 93% PFS at 24 months.15 Inthis study, the nine patients with 17p deletion all experienceda response, including three with complete remissions, andnone have experienced disease progression to date. The pri-mary toxicities with idelalisib have included increase in he-patic transaminases, usually readily manageable byinterrupting the drug, and diarrhea including colitis in a sub-set of patients.The frst registration trial of idelalisib, in which idelalisib

with rituximab was compared to placebo with rituximab inpatients with CLL and signifcant medical comorbiditieswhose disease relapsed within 24 months of prior therapy,was stopped early for effcacy, and presented at the AmericanSociety of Hematology meeting in December 2013.16 110 pa-tients were enrolled on each arm, and 44% had high-risk 17pdeletion. With idelalisib, the ORR was 81% and the hazardratio for PFS was 0.15 (p 0.0001). Overall survival was alsosignifcantly improved in the idelalisib arm,with three deathson that arm compared with nine on the placebo arm (HR0.28, p 0.02). These results have been submitted for FDAconsideration, and two other registration trials in relapsedCLL are ongoing.

IPI-145IPI-145 is the other PI3K inhibitor advanced in clinical trialsin CLL and is a potent inhibitor of the delta as well as thegamma isoforms. The effect of gamma inhibition is unclear,

Table 1. Novel Drugs

Target Drug Mechanism Patients ORR PFS

CD20 Obinutuzumab (G) Type 2 antibody Untreated elderlyCLL3 (n333 G-clbvs 330 R-clb),phase 3

ORR 78% G-clb vs 65%R-clb, pCR 21% vs 7%, P

26.7m G-clb vs 15.2m R-clb,HR 0.39, p0.0001

BTK Ibrutinib Covalent kinase inhibitor(Binds Cys 481)

Rel/refr CLL10 (n85),phase 2

71% ORR 13% PR-L 75% at 26 months

Untreated elderlyCLL11 (n31),phase 2

71% ORR 13% PR-L 96% at 22 months

CC-292 Covalent kinase inhibitor(Binds Cys 481)


50% ORR 13% PR-L, atRP2D 500 mg BID

N.R.

ONO-4059 Covalent kinase inhibitor(Binds Cys 481)


78% ORR 11% PR-L N.R.

PI3K Idelalisib Competitive kinaseinhibitor, PI3K


39% ORR 33% PR-L 29m median at RP2D150 mg BID

Rel/refr CLL16, (n110Idela-R vs 110placebo-R), phase 3

ORR 81% Idela-R vs 13%placebo-R, p

Median not reached, Idela-Rvs 5.5m placebo-R,HR 0.15, p0.001

IPI-145 Competitive kinaseinhibitor, PI3K and


47% ORR 98% nodalresponse rate

N.R.

BCL-2 ABT-199 Small molecule BH3mimetic


84% ORR23% CR

N.R.

Abbreviations: Clb, chlorambucil; R, rituximab; ORR, overall response rate; PFS, progression-free survival; HR, hazard ratio; Cys, cysteine; Rel, relapsed; refr, refractory; PR-L, partial response withlymphocytosis; RP2D, recommended phase II dose; N.R., not reported; PI3K, the delta isoform of phosphatidylinositol 3 kinase; PI3K, the gamma isoform of PI3K; CR, complete response.



as gamma is expressed in CLL cells as well as T cells andneutrophils, with the potential to increase effcacy or toxic-ity. Dose escalation of IPI-145 has been completed with amaximum tolerated dose of 75 mg twice daily.17 In CLL,however, the dose selected to move forward is 25 mg twicedaily, since this dose achieves full delta inhibition, has a sim-ilarly high response rate, and may show less toxicity. At themost recent report, 47 patients with relapsed refractory CLLhave been treated, with a 98% nodal response rate and 47%IWCLL ORR. Follow-up is still quite short, at 5.2 months asof December 2013, suggesting many patients may have per-sistent lymphocytosis at this point. A registration trial com-paring IPI-145 to ofatumumab in relapsed refractory CLLhas been initiated.

BCL-2 INHIBITION: ABT-199BCL-2 is nearly universally overexpressed in CLL, making itan excellent therapeutic target. The frst-generation clinicalinhibitor ABT-263 had a 35% response rate in heavily pre-treated patients with CLL but dosing was limited by throm-bocytopenia that resulted from on-target inhibition of

BCL-XL in platelets.18 ABT-199 is a second-generation in-hibitor that is highly selective for BCL-2 and is currentlycompleting its phase I study.19 Thrombocytopenia has notbeen a problem on this study, in which the primary toxicityhas been tumor lysis syndrome. The dose-escalation schemahas been revised twice to manage the tumor lysis and cur-rently the drug is slowly escalated weekly over 4 weeks, ini-tially in the hospital, to the planned maximum dose in CLL,which is 400 mg daily. ABT-199 has been very effective, with88% nodal response, 100% lymphocyte response, and 89%bone marrow response, generally soon after therapy initia-tion. TheORR is 84%with 23% complete remissions, includ-ing some patients negative for minimal residual disease.Forty-three of 67 patients remain on study at a median of 11months as of December 2013. Registration trials are plannedfor this drug but may be delayed as the extent of requiredmonitoring is clarifed.

THE USE OF CHIMERIC ANTIGENRECEPTORMODIFIED T CELLS TO TREAT CLLPatients withmultiply relapsed and/or refractory CLL have apoor prognosis with few effective treatment options. Autol-

FIG 1. The chimeric antigen receptor consists of the single chain variable fragment of an antibody (scFv) that recognizes the CD19protein on B cells and leukemia cells coupled to the CD3 zeta activation domain and co-stimulatory domains from CD28 and/or 4-1BB.This combines the MHC independent recognition of a tumor antigen with the activating potential of the T-cell receptor, allowing forredirection of T cells to leukemia. Inset shows normal T cell:APC interaction through T cell receptor/CD3 recognition of an antigen andco-stimulation through CD28 or CD137 (41BB). In the foreground a T cell comes into contact with a CD19 tumor cell at white box. Inthe background, magnied illustration of T cell interaction with CD19 on tumor cell through a CAR molecule containing the scFv anti-CD19 recognition fragment and internal costimulation domains. Most modern constructs include the CD3z stimulatory domain (rstgeneration CARs) with the addition of 4-1BB or CD28 (second generation CARs). CARs are now being tested that contain all 3stimulatory domains (3rd generation). Sue Seif. Reproduced with permission from Sue Seif and Stephan Grupp.



ogous T cells engineered to express a chimeric antigen recep-tor (CAR) have shown recent success in the treatment ofadvanced CLL by combining the specifcity of antibody-directed targeting with the extensive in vivo amplifcationand cell-mediated killing characteristic of cellular therapy.The long-term persistence of this genetically engineered bi-ologic agent can then provide ongoing tumor control andvaccine-like protection against recurrence.A CAR is composed of a targeting domain, a hinge region

for flexibility, a transmembrane domain, and intracellularsignaling components. The targeting domain is typically de-rived from a single-chain variable fragment (scFv) of an an-tibody recognizing a cell-surface antigen of choice (CARstypically cannot recognize intracellular antigens). The signal-ing domain is critical for optimal function of the CAR andgenerally includes the CD3 chain alone (frst-generationCAR) or a signaling domain plus a costimulatory domainsuch as CD28 or 41BB (CD137; second-generation CAR).20Second-generation CAR-modifed T cells are more potentthan frst-generation CARs. Third-generation CARs incor-porating more than one costimulatory molecule are in earlystages of clinical testing.21 A schematic of this approach isshown in Figure 1.CD19 is an excellent tumor target for CAR therapy; expres-

sion is limited to B-cell malignancies, normal B cells, and asmall population of other immune cells. It is not expressed onhematopoietic stem cells or other normal tissue.22 Althougheffective anti-CD19 directed therapy therefore also targetsnormal B cells, it is not expected to induce marrow aplasia.Previous limitations to effective cellular therapy included

lack of specifcity of T cells, inability to generate suffcientnumbers of cells for clinical application, minimal in vivo ex-pansion, and poor persistence. Improved technology for in-troduction of new genetic material, and for T-cell expansion,means that large numbers ofmodifedT cells cannowbe gen-erated in 10 to 14days.MostCAR trials have expandedT cellsin high numbers by exposure to anti-CD3/anti-CD28 coatedbeads that can be removed before infusion.23 Several meth-ods exist for the stable introduction of a CAR into T cells, butmost studies have used lentiviral or retroviral transductionwith effciencies consistently greater than 20%. After trans-duction and integration, CAR-modifed T cells are expandedex vivo over approximately 2 weeks. Using these techniques,recent trials have shown that after intravenous infusion,CAR-modifed T cells can undergo remarkable in vivo pro-liferation and persist for long periods of time, overcomingsome of the previous barriers to successful cellular immuno-therapy. Typically patients receive lymphodepleting chemo-therapy before CAR cell infusion with the intent ofpromoting homeostatic proliferation and expansion of theinfused CAR-modifed T cells. It remains unknown, how-ever, whether pre-infusion chemotherapy is actually neces-sary in all patients.At the University of Pennsylvania (UPenn), autologous T

cells modifed with a CAR directed against CD19 that con-tains the CD3/41BB signaling domains (termed CTL019cells) were used to treat patients with relapsed and refractory

CLL. It was initially reported that CTL019 cells could un-dergo massive in vivo proliferation, result in delayed tumorlysis, and induce potent antitumor responses.24,25 The clini-cal trial that included 14 patientswith relapsed and refractoryCLL has been completed26; a schematic of the trials is shownin Fig. 2. Patients were heavily pretreated with a median offour prior regimens. Similar to other trials using anti-CD19CAR, all patients received lymphodepleting chemotherapybefore CTL019 cell infusion. Patients received a median of7.5 x 108 total cells (range: 1.7 to 50 x 108), corresponding to1.4 x 108 (range: 0.14 to 5.9) genetically modifed cells. Inmany cases, these cells were capable of undergoing dramaticproliferation with more than a 3-log increase in vivo anddemonstrated long-term persistence. Four patients (29%)achieved a CR and 4 (29%) achieved a PR for an overall re-sponse rate of 57%. CAR-modifed T cells have been detectedby flow cytometry beyond 3 years after infusion in some pa-tients, correlating with a similar remission duration of morethan 3 years. There was no correlation between response andcell dose administered, patient age, prior therapy, or cytoge-netic risk profle. To better defne the activity of CTL019 cellsand determine the contribution of the cell dose, a random-ized phase II study comparing two different doses of CTL019cells is being performed. Although still in progress, an earlyanalysis on the frst 18 patients treated has not identifed asignifcant dose:response or dose:toxicity relationship.27 Thisis a surprising and important fnding and differs from resultsseen with standard drugs that are not self-replicating; re-sponse and toxicity are more likely determined by the num-ber of cells generated by in vivo proliferation and expansionrather than by the dose infused.Several other studies using gene-modifed T cells to target

CD19 have been reported and have been recently summa-rized.28 Using an anti-CD19 CAR containing the CD28 co-stimulatory domain and retroviral transduction, the group atMemorial Sloan Kettering Cancer Center have reportedmodest activity in at least one of eight patients with CLL.29Remissions in patients with CLL have been shown by thegroup at the National Cancer Institute as well, using a moreintensive lymphodepleting chemotherapy regimen and ananti-CD19 CAR containing the CD28 costimulatory do-main.30 Interestingly, this group has also shown that donor-derived CAR-modifed T cells can be used to treat relapsedCLL after allogeneic stem cell transplantationwithout induc-tion of graft-versus-host disease.31 The major differences be-tween these trials and the studies done at the University ofPennsylvania are the use of retroviral-mediated introductionof the CAR construct and the inclusion of the CD28 co-stimulatory domain rather than 4-1BB. The design of theclinical trials are otherwise quite similar. Whether these dif-ferences account for the variations differences in activity andoutcome is not known at this time.

TOXICITYA number of toxicities can be anticipated with anti-CD19CAR T-cell therapy. B-cell aplasia is an expected on-targetcomplication resulting in prolonged hypogammaglobuline-



mia. Indeed, B-cell aplasia can be considered a biomarker forongoing CAR T-cell activity. Hypogammaglobulinemia isusually managed with regular immunoglobulin repletion;whether immunoglobulin repletion is in fact necessary toprevent infections is not known andwill need to be studied infuture trials.The development of a delayed tumor lysis syndrome

(TLS) is a testament to the proliferative capacity and potencyof CAR T cells and has been seen in several CAR studies.25,31TLS typically develops at the height of T-cell proliferationand is associated with a cytokine release syndrome that re-quires careful monitoring during the frst several weeksafter infusion; a xanthine oxidase inhibitor has been admin-istered for several weeks after T-cell infusion to preventcomplications from TLS. Once established, management ofTLS is generally successful with standard supportive caremeasures.Patients with a disease response almost all develop a cyto-

kine release syndrome (CRS). This is one of the more uniquecomplications of CAR T-cell therapy and has been reportedin varying degrees of severity in most CAR T-cell trials. TheCRS is manifested by symptoms that can include high fevers,rigors, anorexia, nausea, myalgias, and arthralgias, and prog-ress to hypotension, capillary leakwith pulmonary infltrates,and hypoxia. This is believed to be related to cytokine pro-duction from the rapidly expanding T cells, though cytokines

can also be produced by activated macrophages and/ordying tumor cells. The CRS has also been associated with amacrophage activation syndrome (MAS) associated withhemophagocytosis, cytopenias, hypofbrinogenemia, andmarked elevations and ferritin.32 In particular, very high lev-els of interleukin-6 (IL-6) and interferon gamma have beennoted during the CRS. This led to interventions using theIL-6 receptor antagonist tocilizumab, which has consistentlyled to rapid improvement in the clinical and biochemical ab-normalities associated with CRS and MAS.32 In the UPennexperience, the CRS has not been particularly responsive tocorticosteroids, although others have reported successfultreatment of a similar CRS with high doses of corticoste-roids.33 It is not yet known whether blockade of cytokineswith anticytokine therapy such as tocilizumab or corticoste-roids affects the antitumor response. Therefore, interventionfor severe CRS has been effective, but the most appropriatetiming of intervention remains unknown. Treating patientswith lower tumor burdens may produce a less-intense cyto-kine reaction,33 lending support to using this approach be-fore the disease becomes refractory and patients haveextensive disease.There are other theoretical risks to consider when using

integrating retroviral or lentiviral vectors, transposons, andelectroporation to introduce new genetic material into the Tcell. For instance, random integration could result in inser-

FIG 2. Overview of CTL019 therapy. (1) Leukapheresis: the patients own T cells are harvested; (2) T cells are activated ex vivoexpansion on antibody-coated beads and genetically transduced ex vivo with a construct encoding the anti-CD19 chimeric antigenreceptor; (3) CTL019 cells undergo ex vivo expansion on antibody-coated beads; (4) patients receive a preparative lymphodepletingregimen before T-cell infusion; (5) CTL019 cells are reinfused into the patient where they undergo in vivo expansion and target CD19cells for destruction. They remain persistent in the body to guard against residual or recurring disease.



tional mutagenesis and induction of T-cell lymphoprolifera-tive disorders. This risk appears to be low based on the long-term follow-up of patients with human immunodefciencyvirus (HIV) treatedwith a CD4CAR,where there have beenno cases of genotoxicity in more than 540 patient-years ofobservation.34 In addition, the development of replication-competent virus is also highly unlikely as supported by ex-tensive follow-up from 29 different clinical trials.35More than 40 patients with CLL have been treated at

UPenn andmany patients have been treated at other centers.To date, the majority of clinical trials have included patientswith advanced and heavily pretreated CLL. In a short periodof time, investigations have developed a better understand-ing of the potential activity, anticipated short-term tox-icities, and outcomes using autologous T cells geneticallymodifed with an anti-CD19 CAR. It is reasonable to nowconsider testing this approach earlier in the course of disease(when toxicitymay beminimized and before exposure to pu-rine analogs and other chemotherapy agents with the poten-tial for long-term marrow damage and other toxicity) orperhaps at the time of minimal residual disease.36 Movingforward, CARs directed against other CLL-specifc antigenswill be tested and it will be exciting to test CAR-modifed Tcells in combination with other novel agents or immunemodulators thatmay augment or regulate the activity of CART cells.

FUTURE PLANSThe advent of tyrosine kinase inhibitors (TKIs) aswell as newmonoclonal antibodies in the treatment of CLL have raisedthe question of whether chemoimmunotherapy, the currentof standard of care, will soon become a thing of the past. Al-though these novel agents vary in terms of their side effects,the one thing that they have in common is the lack of myelo-suppression. This is a marked contrast to chemoimmuno-therapy such as FCR where the primary initial complicationis myelosuppression and associated infections. Obinutu-zumab, the new anti-CD20 monoclonal antibody, recentlyreceived FDA approval for initial use in CLL in combinationwith chlorambucil. It is likely that ibrutinib and idelalisib willboth be approved by the FDA this year for patients who withrelapsed CLL, and ABT-199 following on thereafter. Al-though the initial label for these drugs will be in relapsed dis-ease, there are front-line trials ongoing, and planned, to allowfor expansion of the label to previously untreated patients. Inaddition, since off-label use of agents is common in theUnited States, particularlywhen there is published data avail-able, this is likely to occur. This raises the question ofwhetherchemotherapy-free regimens should be initiated in the nearfuture, or whether there is still a role for regimens such asFCR.The use of FCR as initial therapy for patients with CLL re-

sulted in a median PFS of approximately 5 years in the Ger-man CLL 8 trial in which patients were randomly assignedbetween FC and FCR.37 Phase II data from theMDAndersonCancer Center single-center experience suggested a PFS of

approximately 6 to 7 years.38 This longer PFS was likely seenbased on extensive local experiencewith this regimen and themore extensive use of growth factor support to facilitate com-pletion of six cycles; growth factor support has not routinelybeen used in the German CLL trials. The only front-line datacurrently available for the new agent involves the use of ibru-tinib in treatment-nave patients older than age 65.11 Thisgroup was specifcally targeted since there is widespreadagreement in the CLL community that themyelosuppressionand infections associated with chemoimmunotherapy aremore problematic in older patients or those with signifcantcomorbidities. In that trial with ibrutinib themedian PFS hasnot been reached, and only one patient has experienced re-lapsed diseasewith amedian follow-up of 30months.39How-ever, it is important to note that this experience is based onanalysis of only 31 patients; although this early data is im-pressive, there is no 5-year follow-up to offer a comparison tofront-line chemoimmunotherapy. Nevertheless, the relativeease of administration of this agent in a patient populationwhich is less able to complete a full course of chemoimmu-notherapy suggests that there will be rapid adoption of TKIsin this population.The other population in which chemoimmunotherapy is

relatively ineffective are patients with 17p deletion. It is wellknown that these patients have a poor disease response tochemotherapy-based regimens, partly related to the absenceof p53. Themedian PFS of patients with 17p deletions treatedwith FCR as front-line therapywas approximately 12monthsin the MD Anderson Cancer Center analysis38 and identicalin the German CLL 8 trial.37 Clearly, a 12-month PFS in afront-line population is poor, and this is another group thatwill rapidly be treated with novel agents rather than chemo-immunotherapy. Data from the ibrutinib trial in patientswith relapsed and refractory CLL indicated that the medianPFS of patients relapsed/refractory disease with 17p deletionwas 2 years, signifcantly longer than the front-line PFS seenwith chemoimmunotherapy.Recent data from long-term follow-up of a cohort of 300

patients fromMDAnderson Cancer Center presented at theiwCLL meeting in 2013 showed that approximately 30% ofthese patientswere still in remission at 10 to 14 years from thestart of chemotherapy. Although 30% is a minority of pa-tients, if this group could be reliably identifed, then perhapsin those patients continued use of chemoimmunotherapy insome form is a strategy that should be retained.An analysis ofprognostic factors suggested that such patients can be iden-tifed, they were patients who did not have a 17p or 11q dele-tion, had amutated IgVH gene and frequently had trisomy 12or 13q deletion. In fact, when the PFS analysis was limitedonly to patients with a mutated IgVH gene, 60% were still inremission at 10 years.However, late complications of chemoimmunotherapy

have been recognized, and there may be some that have notyet been reported. A serious, albeit infrequent, complicationis the development of treatment-relatedmyelodysplastic syn-drome (t-MDS) or acute myelogenous leukemia (AML).38This is a devastating phenomenon because treatment out-



comes are poor for these secondary malignancies, and sur-vivals are generally short. This raises the question of whetherthe use of novel antibodies such as obinutuzumab or oral in-hibitors in combination with chemotherapy may allow theduration of the regimen (usually six cycles/6 months) to beshortened and produce the same or better outcomes withlimited exposure to chemotherapy.The low CR rate seen even in the trial in patients receiving

front-line treatment may suggest that 10-year remissions areunlikely to occur with single-agent TKIs. However, given theshort follow-up in the front-line population and the contin-ued evolution of responses over time, it is possible that in thatgroup CR rates will increase beyond the currently seen 13%.Although responses may continue to improve in the re-lapsed/refractory population, clearly the constant decline inthe PFS and OS curves in these patients suggests that thecompeting risk of resistance will likely not result in a largeincrease in the CR rate. Data showing that mutations in bothBTK and PLC gammamay occur in resistant disease recapit-ulate fndings seen with TKIs in the setting of chronic my-elogenous leukemia, namely that specifc targeting of kinasesinevitably results in a proportion of patients developing re-sistance through mutations of the target.40One obvious question is whether new TKIs or BCL-2 in-

hibitors can be combined with each other or with novel an-tibodies such as obinutuzumab. At this time there is noclinical trial data to answer this, but there is a reasonable ra-tionale for combining various agents. There is some data

evaluating TKIs in combination with antibodies, clearlyshowing that the use of rituximab or ofatumumab results infaster responses because of the abrogation of lymphocytosisseen with single-agent TKIs. However, current data, admit-tedly with very short follow-up, suggest the CR rates are notincreased and PFS may not be signifcantly enhanced in thissetting.41 Combining oral targeted agents is a very attractiveconcept but raises an important and ever more present issuein cancer therapeutics, which is the cost of drugs. These novelagents are likely to be very expensive and combining themoutside of a clinical trial may be prohibitory. In addition, be-cause of the current reimbursement schema for oral thera-peutics requiring a copay (which varies signifcantly frominsurance to insurance), even when these drugs are coveredby insurance the concomitant copay may make use of theagent impossible in a givenpatient. Taking a lesson fromHIVdrugs would suggest that if such combinations could beachieved in a single pill this strategy would be more practicalin the long run.Currently, it is very exciting to have these novel, relatively

nontoxic, oral agents available that are not myelosuppressiveand can be easily administered. Although chemotherapy isassociated with the short-term complications of myelosup-pression and infection as well as the low risk of late t-MDS ort-AML, the therapy is inexpensive and treatment is fnishedafter 6months. The rapidly rising cost of cancer care suggeststhat this is something that will have to be factored into thealgorithm moving forward.

Disclosures of Potential Conicts of Interest

Relationships are considered self-held and compensated unless otherwise noted. Relationships marked L indicate leadership positions. Relationships marked I are those held by an immediatefamily member; those marked B are held by the author and an immediate family member. Relationships marked U are uncompensated.

Employment or Leadership Position: David L. Porter, Genentech/Roche (I). Consultant or Advisory Role: Jennifer R. Brown, Boehringer Ingelheim;Celgene; Emergent; Genentech; Novartis; Onyx; Pharmacyclics; Sano; Vertex. Susan M. OBrien, Amgen; Celgene; Emergent BioSolutions; Genentech; Genmab;Gilead Sciences; GlaxoSmithKline; Innity; Lilly; MorphoSys; Pharmacyclics; Sunesis Pharmaceuticals; Talon Therapeutics; Teva. Stock Ownership: None.Honoraria: None. Research Funding: Jennifer R. Brown, Celgene; Genzyme.Susan M. OBrien, ARIAD; Bayer; Biogen Idec; Bristol-Myers Squibb; CalistogaPharmaceuticals; Emergent BioSolutions; Gemin X; Genentech; Genta; Gilead Sciences; Hana BioSciences; Innity; Lilly; MorphoSys; Novartis; Pharmacyclics;Talon Therapeutics. David L. Porter, Novartis. Expert Testimony: None. Other Remuneration: None.

References

1. Alduaij W, Ivanov A, Honeychurch J, et al. Novel type II anti-CD20monoclonal antibody (GA101) evokes homotypic adhesion and actin-dependent, lysosome-mediated cell death in B-cell malignancies. Blood.2011;117:4519-4529.

2. Goede V, Fischer K, Busch R, et al. Chemoimmunotherapy with GA101plus chlorambucil in patients with chronic lymphocytic leukemia andcomorbidity: results of the CLL11 (BO21004) safety run-in. Leukemia.2013;27:1172-1174.

3. Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil inpatients with CLL and coexisting conditions. N Engl J Med. Epub 2014Jan 8.

4. Landau DA, Carter SL, Stojanov P, et al. Evolution and impact of sub-

clonal mutations in chronic lymphocytic leukemia. Cell. 2013;152:714-726.

5. ChesonBD, Byrd JC, Rai KR, et al. Novel targeted agents and the need torefne clinical end points in chronic lymphocytic leukemia. J Clin Oncol.2012;30:2820-2822.

6. Advani RH, Buggy JJ, Sharman JP, et al. Bruton tyrosine kinase inhibitoribrutinib (PCI-32765) has signifcant activity in patients with relapsed/refractory B-cell malignancies. J Clin Oncol. 2013;31:88-94.

7. Honigberg LA, Smith AM, Sirisawad M, et al. The Bruton tyrosine ki-nase inhibitor PCI-32765 blocks B-cell activation and is effcacious inmodels of autoimmune disease and B-cell malignancy. Proc Natl AcadSci U S A. 2010;107:13075-13080.



8. Dubovsky JA, Beckwith KA, Natarajan G, et al. Ibrutinib is an irrevers-ible molecular inhibitor of ITK driving a Th1-selective pressure in Tlymphocytes. Blood. 2013;122:2539-2549.

9. Chang BY, Furman RR, Zapatka M, et al. Use of tumor genomic profl-ing to reveal mechanisms of resistance to the BTK inhibitor ibrutinib inchronic lymphocytic leukemia (CLL). J ClinOncol. 2013;31 (suppl; abstr7014).

10. Byrd JC, Furman RR, Coutre SE, et al. Targeting BTK with ibrutinib inrelapsed chronic lymphocytic leukemia. N Engl J Med. 2013;369:32-42.

11. OBrien S, Furman RR, Coutre SE, et al. Ibrutinib as initial therapy forelderly patients with chronic lymphocytic leukaemia or small lympho-cytic lymphoma: an open-label, multicentre, phase 1b/2 trial. LancetOncol. 2014;15:48-58.

12. Brown JR,HarbWA,Hill BT, et al. Phase 1 study of single agentCC-292,a highly selective Brutons tyrosine kinase (BTK) inhibitor, in relapsed/refractory chronic lymphocytic leukemia (CLL).Blood. 2013;122 (suppl;abstr 1630).

13. Salles GA, Karlin L, Rule S, et al. A phase I study of the oral Btk inhibitorONO-4059 in patients with relapsed/refractory and high risk chroniclymphocytic leukaemia (CLL). Blood. 2013;122 (suppl; abstr 676).

14. Brown JR, Furman RR, Flinn I, et al. Final results of a phase I study ofidelalisib (GS-1101) a selective inhibitor of PI3K, in patients with re-lapsed or refractory CLL. J Clin Oncol. 2013;31 (suppl; abstr 7003).

15. OBrien SM, LamannaN, Kipps TJ, et al. A phase II study of the selectivephosphatidylinositol 3-kinase delta (PI3K) inhibitor idelalisib (GS-1101) in combination with rituximab (R) in treatment-naive patients(pts)65 years with chronic lymphocytic leukemia (CLL) or small lym-phocytic lymphoma (SLL). J Clin Oncol. 2013;31 (suppl; abstr 7005).

16. Furman RR, Sharman JP, Coutre SE, et al. A phase 3, randomized,double-blind, placebo-controlled study evaluating the effcacy andsafety of idelalisib and rituximab for previously treated patients withchronic lymphocytic leukemia (CLL). Blood. 2013;122 (suppl; abstrLBA-6).

17. Flinn I, Patel M, Kahl BS, et al. Preliminary safety and effcacy of IPI-145, a potent inhibitor of phosphoinositide-3-kinase-,, in patientswith chronic lymphocytic leukemia. Blood. 2013;122 (suppl; abstr 677).

18. Roberts AW, Seymour JF, Brown JR, et al. Substantial susceptibility ofchronic lymphocytic leukemia to BCL2 inhibition: results of a phase Istudy of navitoclax in patients with relapsed or refractory disease. J ClinOncol. 2012;30:488-496.

19. Seymour JF,DavidsMS, Pagel JM, et al. Bcl-2 inhibitorABT-199 (GDC-0199) monotherapy shows anti-tumor activity including complete re-missions in high-risk relapsed/refractory (R/R) chronic lymphocyticleukemia (CLL) and small lymphocytic lymphoma (SLL). Blood. 2013;122 (suppl; abstr 872).

20. Milone MC, Fish JD, Carpenito C, et al. Chimeric receptors containingCD137 signal transduction domains mediate enhanced survival of Tcells and increased antileukemic effcacy in vivo. Mol Ther. 2009;17:1453-1464.

21. Till BG, JensenMC,Wang J, et al. CD20-specifc adoptive immunother-apy for lymphoma using a chimeric antigen receptor with both CD28and 4-1BB domains: pilot clinical trial results. Blood. 2012;119:3940-3950.

22. Scheuermann RH, Racila E. CD19 antigen in leukemia and lymphomadiagnosis and immunotherapy. Leuk Lymphoma. 1995;18:385-397.

23. Levine BL, BernsteinWB, Connors M, et al. Effects of CD28 costimula-tion on long-term proliferation of CD4 T cells in the absence of exog-enous feeder cells. J Immunol. 1997;159:5921-5930.

24. Kalos M, Levine BL, Porter DL, et al. T cells with chimeric antigen re-ceptors have potent antitumor effects and can establish memory in pa-tients with advanced leukemia. Sci Transl Med. 2011;3:95ra73.

25. Porter DL, Levine BL, Kalos M, et al. Chimeric antigen receptor-modifed T cells in chronic lymphoid leukemia.NEngl JMed. 2011;365:725-733.

26. PorterD, KalosM, FreyNV, et al. Chimeric antigen receptormodifed Tcells directed against CD19 (CTL019 cells) have long-term persistenceand induce durable responses in relapsed, refractory CLL. Blood. 2013;122 (suppl; abstr 4162).

27. Porter D, Kalos M, Frey NV, et al. Randomized, phase II dose optimi-zation study of chimeric antigen receptor modifed T cells directedagainst CD19 (CTL019) in patients with relapsed, refractory CLL.Blood. 2013;122 (suppl; abstr 873).

28. Maus M, Grupp SA, Porter D, et al. CAR-modifed T cells for hemato-logic malignancies. Blood. In press.

29. Brentjens RJ, Rivie`re I, Park JH, et al. Safety and persistence of adop-tively transferred autologous CD19-targeted T cells in patients with re-lapsed or chemotherapy refractory B-cell leukemias. Blood. 2011;118:4817-4828.

30. Kochenderfer JN, Dudley ME, Feldman SA, et al. B-cell depletion andremissions of malignancy along with cytokine-associated toxicity in aclinical trial of anti-CD19 chimeric-antigen-receptor-transduced Tcells. Blood. 2012;119:2709-2720.

31. Kochenderfer JN, Dudley ME, Carpenter RO, et al. Donor-derivedCD19-targeted T cells cause regression of malignancy persisting afterallogeneic hematopoietic stem cell transplantation. Blood. 2013;122:4129-4139.

32. Grupp SA,KalosM,BarrettD, et al. Chimeric antigen receptor-modifedTcells for acute lymphoid leukemia.NEngl J Med. 2013;368:1509-1518.

33. Brentjens RJ, Davila ML, Rivie`re I, et al. CD19-targeted T cells rapidlyinduce molecular remissions in adults with chemotherapy-refractoryacute lymphoblastic leukemia. Sci Transl Med. 2013;5:177ra38.

34. Scholler J, Brady TL, Binder-Scholl G, et al. Decade-long safety andfunction of retroviral-modifed chimeric antigen receptor T cells. SciTransl Med. 2012;4:132ra53.

35. Bear AS, Morgan RA, Cornetta K, et al. Replication-competent retrovi-ruses in gene-modifed T cells used in clinical trials: is it time to revisethe testing requirements?Mol Ther. 2012;20:246-249.

36. Park JH, Rivie`re I, Wang X, et al. Phase I trial of autologous CD19-targeted CAR-modifed T cells as consolidation after purine analog-based frst-line therapy in patients with previously untreated CLL.Blood. 2013;122 (suppl; abstr 874).

37. HallekM, Fischer K, Fingerle-RowsonG, et al. Addition of rituximab tofludarabine and cyclophosphamide in patients with chronic lympho-cytic leukaemia: a randomised, open-label, phase 3 trial. Lancet. 2010;376:1164-1174.

38. Tam CS, OBrien S, Wierda W, et al. Long-term results of the fludara-bine, cyclophosphamide, and rituximab regimen as initial therapy ofchronic lymphocytic leukemia. Blood. 2008;112:975-980.

39. OBrien S, Furman RR, Fowler N, et al. The Brutons tyrosine kinase(BTK) inhibitor ibrutinib (PCI-32765) monotherapy demonstrateslong-term safety and durability of response in chronic lymphocytic leu-kemia (CLL)/small lymphocytic lymphoma (SLL) patients in an open-label extension study. Blood. 2013;122 (suppl; abstr 4163).

40. Stilgenbauer S, Chang B, et al. Tumor genomic profling reveals mech-anisms of resistance to BTK inhibitor ibrutinib in chronic lymphocyticleukemia (CLL). iWCLL Sept. 2013.

41. Burger JA,KeatingMJ,WierdaWG, et al. Ibrutinib in combinationwithrituximab (iR) is well tolerated and induces a high rate of durable remis-sions in patients with high-risk chronic lymphocytic leukemia (CLL):new, updated results of a phase II trial in 40 patients. Blood. 2013;122(suppl; abstr 675).



Brown 2014

Documents

Transcript of Brown 2014