`
`
`
`
`NOVEMBER 2000
`
`
`VOL 14 0 NO11A 0 NCCN PROCEEDINGS
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`emmoa
`
`NCCN®
`
`Oncology Practice Guidelines
`
`Volume 7
`
`NCCN Practice Guidelines for
`
`Acute Myelogenous Leukemia
`Chaired by
`Margaret R. O’Donnell, MD
`City of Hope
`National Medical Center
`Breast Cancer
`Chaired by Robert W. Carlson, MD
`Stanford Hospital and Clinics
`Cancer Pain
`Chaired by Stuart A. Grossman, MD
`Johns Hopkins Oncology Center
`Cancer-Related Fatigue
`Chaired by Victoria Mock. DNSC, RN
`Johns Hopkins Oncology Center
`
`Chronic Myelogenous
`Leukemia
`Chaired by Moshe Talpaz. MD
`The University of Texas
`M. D. Anderson Cancer Center
`Colorectal Cancer
`Chaired by Paul F. Engstrom, MD
`Fox Chase Cancer Center
`Head and Neck Cancers
`Panel chaired by
`Arlene A. Forastiere, MD
`Johns Hopkins Oncology Center
`Conference presentation by
`David Pt‘ister, MD
`Memorial Sloan—Kettering
`Cancer Center
`
`
`
`
`
`
`
`
`“IlllllIllllllllllllll”illll'llllllllllllll'
`******/\UTO**MIXEDADC59
`
`ll.l..
`
`*thfi'
`
`Ctr
`
`L05AngéflesCA90095-1798
`
`
`
`
`
`
`
`~SeriulsHoa1thServices
`
`
`CDLIBC(rriher[798
`
`
`CMPUCLABIOM12-077Box95
` For ONCOLOGY on the'Web, visit www.cancernetwork.com
`
`
`Prostate Cancer
`Chaired by Laurence H. Baker, Do
`University of Michigan
`Comprehensive Cancer Center
`
`Full table afcoments 021 pages 17-18 and 21-23
`
`|PR2018—01714
`
`Celgene Ex. 2007, Page 1
`
`IPR2018-01714
`Celgene Ex. 2007, Page 1
`
`

`

`
`
`
`
`MAYER N. FISHMAN, MD, PhD
`Assistant Professor
`
`Interdisciplinary Oncology Program
`
`WILLIAM S. DALTON, MD, PhD
`Professor and Chairman
`H. Lee Moffitt Cancer Center
`& Research Institute
`
`University of South Florida
`Tampa, Flei-ida
`
`Considerations in
`the Management
`of Myeloma
`
`ore than 13,500 cases of mul—
`tiple myeloma will have been
`diagnosed in the United States
`in 2000. Treatments are designed to pro-
`long the symptom-free interval, overall
`survival. and quality of life. Despite the
`development of numerous treatment reg—
`imens, median survival remains less
`than 4 years. For a few patients, howev-
`er, potentially curative therapy exists.
`We will first review issues of diag—
`nosis, estimation of prognosis, and mea—
`surement of response to therapy. The
`latter part of the article will address
`contemporary management options,
`with an emphasis on the ongoing clini—
`cal development of newer transplant
`approaches and drug applications.
`When discussing prognosis and
`treatment options with patients, the phy—
`sician should address the following ba—
`sic issues: (1) What distinguishes the
`diagnosis of myeloma from other mon—
`oclonal gammopathies‘? (2) When in
`the disease course is treatment neces—
`
`sary? (3) Which drugs should be used
`for treatment? (4) How should non—
`cytotoxic drugs,
`including interferon
`biSphosphonates, and growth factors be
`integrated? (5)
`Is high-dose therapy
`with stem—cell rescue or allogeneic bone
`marrow transplantation appropriate?
`Understanding these issues will be use—
`ful when planning a consolidated ap-
`proach for
`all phases of
`the
`disease—from initial
`treatment and
`
`maintenance to salvage therapy and
`palliation. We are hopeful that these
`guidelines will provide a starting point
`from which risks and benefits can be
`individualized.
`
`aBSTntqr
`
`
`
`Multiple myeloma remains an incurable cancer. In recentyears, progress
`in different drug classes has improved outcomes, but management has
`become more complicated. Areas such as prognostic classification, the
`increased use ofhigh-dose chemotherapy with autologousstem-cell rescue,
`and a wider array of ancillary drugs must be integrated into recommenda-
`tionsfora consolidated treatmentplan. Estimatingprognosis is dependent
`on both clinicalfeatures and a growing list of laboramry tests. Autologous
`transplantation has been applied to an increasingproportion ofpatients, at
`different points in the disease process. Besides the age cut-off issue, there
`are still significant treatment choices to be made within the transplant
`technique. Newer drugs, most recently, thalidornide (Thalomid), may offer
`benefits independent of conventional cytotoxic drugs or steroids. Use of
`ancillary drugs, such as bisphosphonates, interferon, P—glycoprotein block-
`ers, antibiotics, and growth factors, are also discussed. For the future,
`immunotherapy in the posth‘ansplantsetting appearspromising. Ultimate-
`ly, basic research must identify intmcellular targetsfor the development of
`specific new-generation drugs.
`
`Diagnosis
`
`A referral for diagnosis of myeloma
`may result from abnormalities on rou-
`tine tests or from a presentation with
`symptoms. Test abnormalities may oc-
`cur at any stage of disease, but a pre-
`sentation of symptoms is usually
`indicative of stage III disease. The phy-
`sician must decide not only how to treat
`the patient, but also when.
`Laboratory results, such as those
`showing anemia, hypeiproteinemia, re-
`nal failure, or hypercalcemia, may ex-
`plain the symptoms. Alternatively,
`infection, neurologic symptoms, abnor—
`
`mal bone imaging, or pathologic frac—
`ture, may lead more indirectly to the
`diagnosis. The need to improve symp—
`toms may obviate the decision regard-
`ing when to proceed with treatment.
`However, more commonly, the deci—
`sion will be based on the factors dis-
`cussed below.
`
`The initial parts of the diagnostic
`work-up algorithm from the 1998 Na—
`tional Comprehensive Cancer Net-
`work‘s (NCCN) Guidelines are in Table
`l.[1] The major and minor diagnostic
`criteria of Durie and Salmon are repro~
`duced in Table 2.[2] Examination of a
`unilateral marrow aspirate and biOpsy
`
`|PR2018-01714
`72
`ONCOLOGY - VOLUME 14 - \lUMBER 11A - NCCN PROCEEDINGS
`
`Celgene Ex. 2007, Page?
`
`'0'
`
`IPR2018-01714
`Celgene Ex. 2007, Page 2
`
`

`

`
`
`Waldenstrom’s macroglobulinemia,
`While there is,
`in some cases, an
`apparent functional overlap of the ther—
`apeutic optiOns (eg, alkylators, steroids
`for Waldenstrom’s macroglobulinemia),
`distinct treatments are usually recom-
`mended. For example, a radiation dose
`of 40 cGy to 55 cGy would be recom—
`mended for plasmacytoma, compared
`to < 30 cGy for palliation of a symp-
`tomatic myeloma lesion.[l] The diag-
`nosis should be clear before proceeding
`to treatment.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`If the patient presents with MGUS
`or smoldering myeloma, a serial obser—
`vation will be necessary to rule out pro—
`gressive disease. Diagnostic criteria for
`MGUS, indolent myeloma, or smolder-
`ing myeloma (also described by Durie
`and Salmon) are in Table 3.[2] Solitary
`plasmacytomas are distinguishable by
`having noninvolved marrow findings
`away from the single site. Patients with
`solitary plasmacytoma of the bone will
`frequently convert to multiple myeloe
`ma and require long—term follow—up.
`Waldenstr'om’s macroglobulinemia—an
`infrequent and indolent disorder—is dis—
`tinguished principally by the immuno-
`globulin M isotype paraprotein, a more
`lymphomatoid appearance of the ma-
`lignant plasma cells. and a clinical
`course similar to low-grade NHL.
`
`
`
`
`
`
`
`
`
`
`
`
`Prognostic Factors
`
`—_—I~———
`
`Table 1
`
`NCCN 1998 Initial Diagnostic
`Guidelines for Multiple
`Myeloma[1]
`
`‘
`
`Diagnostic Work-Up
`'
`H & P
`- CBC
`- Calcium, albumin
`- Quantitative immunoglobulin
`0 SPEP and immunofixation
`- UPEP and immunotixation
`- Quantitation of M protein
`-
`Skeletal survey
`- Unilateral bone marrow aspirate
`and biopsy
`
`Generally Useful
`
`B-2M
`
`Labeling index (PCLI)
`C-reactive protein
`LDH
`
`Useful Under Some Circumstances
`
`MRI for cord compression
`- MRI for suspicion of solitary bone
`plasmacytoma
`- CT to evaluate suspected
`metastases
`
`- Tissue biopsy to diagnose a solitary
`oseeous or extraosseous plasma-
`cytoma
`- Cytogenetios in candidates for
`autologous stem-cell transplantation
`————______
`
`B-2M = betae-mioroglobulin; CBC = complete
`blood count: CT = computed tomography; H&P =
`history and physical; LDH = lactate dehydrogena-
`se; MRI = magnetic resonance imaging: NCCN =
`National Comprehensive Cancer Network; PCLI =
`plasma cell labeling index; SPEP = serum protein
`electrophoresis; UPEP = urinary protein electro-
`phoresls.
`
`is the cornerstOne of pathologic confir-
`mation. Evaluation of a paraprotein is a
`frequent starting point, but light chain
`disease, immunoglobulin D, and non-
`secretory myeloma diagnoses may be
`confirmed in the absence of a detect-
`able paraproteins. Non—Hodgkin’s lym~
`phoma (NHL), chronic lymphocytic
`leukemia, and nonmalignant plasma—cell
`disorders may also have monoclonal
`paraproteins.[3,4] Smoldering or indo-
`lent myeloma, which may be managed
`initially with observation, should be con—
`sidered before proceeding to treatment.
`Similarly, the following non—myeloma
`plasma—cell dyscrasias may bear con—
`sideration: monoclonal gammopathy of
`unknown significance (MGUS), plas—
`macytoma (bone or soft
`tissue), and
`
`
`
`
`
`
`
`
`
`.'
`
`.
`
`
`
`The clinical staging system of Durie
`and Salmon, shown in Table 4,[2,3] is a
`usual starting point for treatment deci—
`sions and prognostic stratification. In
`the 25 years since its publication, addi-
`tional factors to predict prognosis have
`been identified in multiple studies.
`These additional factors reflect tumor
`bulk. growth rate, biology, drug re—
`sponse, and organ—system reserve. The
`most frequently identified factor is the
`serum level of betag-microglobulin
`(B2—M).L5] New prognostic factors may
`often turn out to be closely con-elated
`with previously identified factors, es-
`pecially BZ—M. A list of these prognos—
`tic factors, which are only partly
`evaluated for interdependence,
`is in
`Table 5. Clinical factors. such as stage
`and length of initial plateau phase,[6]
`remain as important as newer molecu-
`lar factors.
`
`Prognostic Stratification
`Prognostic stratification serves two
`
`
`
`
`
`
`
`
`
`
`
`
`purposes: (1) For the individual physi—
`cian/patient relatiOnship, the quantita—
`tion ofthe risk of rapid progression will
`give the patient a more precise estimate
`of prognosis and provide a useful basis
`for making treatment choices. (2) An-
`other pmpose is to achieve more bal—
`ance in the stratification of randomized
`trials or in comparing treatments that
`are described in separate, nonrandom—
`ized studies. Analyses of prognostic fac—
`tors that are continuous variables may
`be facilitated through the use of thresh-
`old values. However, while thresholds
`are useful for group comparisons, it may
`be intuitively unclear how to apply a
`threshold to an individual patient.
`The relative prognostic importance
`of pretreatment factors (especially B2—
`M) can be compared to the importance
`of the assessment of treatment deci—
`sions. Studies[7—10] have consistently
`shown that biologic disease factors ap-
`pear to be more important in predicting
`survival than the treatment decisions
`
`NCCN PROCEEDINGS - NOVEMBER 2000 - ONCOLOGY
`
`73
`
`|PR2018—01714
`Celgene Ex. 2007, Page 31
`
`1
`
`Table 2
`
`Major and Minor Diagnostic
`Criteria
`
`(Multiple myeloma :1 major + 1 minor,
`or 3 minor)
`
`Major
`
`-
`
`Plasmacytoma on tissue biopsy
`
`- Marrow plasmacytosis 30+%
`
`- Monoclonal protein (one of):
`'
`lgG > 3.5
`
`-
`
`lgA > 2
`
`- Bence-Jones >1 924 hours
`Minor
`
`0 Marrow plasmocytosis 10% to 29%
`
`- Monoclonal protein, at less than above
`levels
`
`-
`
`Lytlc bone lesions
`
`- Decrease of the uninvolved
`immunoglobulins
`
`-
`
`-
`
`lgM < 50 mg/dL
`
`lgA < 100 mg/dL
`
`lgG < 600 mg/dL
`-
`——-—-——____
`
`lgA = immunoglobulin A: IgG = immunoglobulin G;
`lgM = immunoglobulin M.
`
`
`
`
`IPR2018-01714
`Celgene Ex. 2007, Page 3
`
`

`

`—__1__
`
`
`
`
`
`Table 3
`
`Table 4
`
`Non-Myeloma Diagnostic
`Criteria
`
`-
`
`-
`
`-
`
`-
`
`Indolent Myeloma
`
`3 or fewer lytic bone lesions
`
`lgA paraprotein < 50 g/L
`
`lgG paraprotein < 70 g/L
`
`- No symptoms
`
`- No anemia <10 g/L
`- Normal calcium
`
`- Normal creatinlne
`
`Smoldering Myeloma
`
`-
`
`Indolent myeloma criteria and
`10% to 30%
`
`- Marrow plasma cells
`- No bone lesions
`
`MGUS
`
`-
`
`-
`
`lgG paraprotein < 35 g/L
`
`lgA paraprotein < 20 g/L
`
`- Bence-Jones protein < 1 9/24 hrs
`
`- < 10% marrow plasma cells
`
`- No symptoms
`- No bone lesions
`
`
`Durie-Salmon Staging System
`
`. A = BUN < 3 mg/dL; creatinine
`< 2 mg/dL
`
`-
`
`B = BUN > 3 mg/dL; creatinine
`> 2 mg/dL
`
`Stage I
`
`.
`
`Low tumor mass (< 0.6 x 10‘2/m3)
`
`All of
`
`- Hgb > 10 g/dL
`
`-
`
`196 < 5 g/dL; lgA < 3 g/dL',
`Benoe-Jones < 4 9124 hours
`- Ca: Normal
`
`'
`
`0 or1 lytic bone lesion
`
`Stage II
`
`-
`
`Intermediate tumor mass
`(0.6 to 1.2 x10'2/m2) neither 1 nor lll
`
`Stage III
`
`' High tumor mass (>12 x “ye/ma)
`
`Any of
`
`- Hgb < 8.5
`
`-
`
`-
`
`lgG > 7 g/dL
`
`lgA > 5 g/dL
`
`- Benoe—Jones >12 9/24 hours
`
`lgA = immunoglobulin A; 196 = immunoglobulin G;
`MGUS = monoclonal gammopathy of unknown
`significance,
`
`- Ca >12 mg/dL (adjusted for albumin)
`
`- Multiple lytio lesions
`
`
`being evaluated, Two polarized views
`may arise from implications of this hi—
`erarchy: (1) One view is that pair—match—
`ing or retrospective estimates do
`accurately gauge whether a new treat—
`ment causes an improvement of out—
`comes or whether the new treatment
`
`produces results that appear better mere-
`ly as a reflection of the selection of a
`patient cohort with better prognostic fea-
`tures,[ 10-12] (2) The other view is that
`only prospective randomization pro—
`vides a fair balance of known molecu-
`
`lar prognostic features, performance
`status, disease stage, lead time, comor—
`bidities, and available supportive
`care.[l3] An unfortunate result of these
`views is that issues of selection bias
`
`and risk stratification may dominate
`comparative discussions of either retro—
`spective or randomized trials. This is
`
`BUN = blood urea nitrogen; Ca = calcium: Hgb =
`hemoglobin; lgA = immunoglobulin A; lgG = im—
`munoglobulin G.
`
`mentioned again below in relation to
`the phase 11 experience with autologous
`transplantation.
`
`Measuring Response
`The finding that the serum or urine
`paraprotein level is directly correlated
`with tumor burden allows for serial
`
`measurements and determination of pro—
`gressive disease and treatment response.
`Progressive disease can be defined as a
`sustained > 25% rise of M protein, er
`the appearance of new bone lesions.[l ]
`Table 6 provides a hierarchy of response
`categories.
`The recommended frequency for
`quantitation of immunoglobulin is ev-
`ery other cycle of therapy, or every 3 to
`
`
`
`
`
`6 months in the observatiori or plateau
`phase of treatment.[ 1] The proposed tar—
`get for quantitative myeloma tumor re-
`duction has been honed because of dose
`intensification and new measurement
`
`techniques.
`The question ofwhether improving
`the frequency of complete response or
`partial reSponse will necessarily improve
`overall survival and event-free survival
`
`must be addressed empirically. For con-
`ventional therapy, a complete response
`does not show an advantage over a par-
`tial response for overall survival, al-
`though plateau duration does influence
`overall survival.[6]
`A higher frequency of complete re—
`sponse and partial response occurs in
`autologous transplants than with con-
`ventiOnal therapy. In the analysis of pa—
`tients treated with up-front,
`tandem
`autologous transplant (see Arkansas
`Group‘s Total Therapy discussed later
`in this article), Barlogie et al found that
`achievement of a complete response, as
`opposed to a partial response, before
`the second tranSplant, resulted in an
`improved median survival (80+ vs 68
`months, P : .001).[14]
`With the advent of molecular tech-
`
`the category of complete re—
`niques,
`sponse (ie, histologically absent
`malignant marrow infiltrate and disap-
`pearance of paraprotein) has been re-
`fined. The molecular complete response
`is the subset of the clinical complete
`response, in which the malignant clone
`is not detected by sensitive polymerase
`chain reaction techniques. Based on the
`experience with other malignancies, a
`molecular complete response may be
`viewed as the rational prerequisite for
`potential cure.[ 15] Early analyses favor
`a better outcome (later relapses) for the
`subset of patients with a molecular com-
`plete response.[l 6]
`Careful, empiric assessment of how
`these better complete responses imply
`or cause improvement of event-free sure
`vival and overall survival is necessary
`in the context of previous experience
`with conventional therapy. Conclusive
`proof of cure among patients achieving
`
`
`Address all correspondence to:
`William S. Dalton. PhD, MD
`H. Lee Moffitt Cancer Center
`& Research Institute
`University of South Florida
`12909 Magnolia Drive
`Tampa. FL 33612
`
`
`
`74
`
`ONCOLOGY - VOLUME 14 - NUMBER 11A - NCCN PROCEEDINGS
`
`|PR2018-01714
`_
`
`Celgene Ex. 2007, Pagejl‘m;
`
`IPR2018-01714
`Celgene Ex. 2007, Page 4
`
`

`

`Table 5
`
`Table 6
`
`a molecular complete response (not just
`prolonged event—free survival) may de—
`velOp in the coming years.
`
`Treatment
`
`
`
`
`
`
`
`
`
`
`Conventional Cytotoxics
`Conventional chemotherapy can be
`divided into alkylator—based (usually
`oral) melphalan (Alkeran) and pred—
`nisone (MP), and non—alkylator—based
`(such as 96—hour continuous infusion)
`vincristine, doxorubicin (Adriamycin),
`and dexamethasone (VAD). The high
`therapeutic index of steroids favors their
`inclusion in most regimens. Single-agent
`dexamethasone has activity without the
`side effects of cytotoxics.
`Numerous published series over the
`last decades have compared different
`conventional therapy arms with a vari-
`ety of alkylator, nitrosourea, steroid,
`vinca, and anthracycline cembina—
`tions.[3,4] A 1992 meta—analysis
`showed that MP appears to be as good
`as other, more complex, toxic, and ex-
`pensive regimens.[ 1 7] More recent anal-
`yses
`have
`reached
`a
`similar
`conclusion.[18,l9]
`Therapy with VAD Offers the fea-
`tures of a more rapid response, without
`the use of alkylating agents, which may
`be toxic to stem cells. Even so, a series
`of 66 patients with stem cells collected
`at the point of salvage had only a 3%
`failure of stem—cell collection.[20] Like
`VAD, high—dose cyclophOSphamide
`(Cytoxan, Neosar)—which also mobi-
`lizes stem cells—is frequently used in
`the pretransplant context. For salvage
`treatment, a non—cross—resistant
`regi—
`men, such as etoposide, dexamethasone,
`ara—C, cisplatin (Platinol) (EDAP),
`VAD; or high—dose cyclophosphamide,
`may be used.[1,15]
`
` -,.-_.
`
`.V-gq.
`
`
`
`_..—.,1~.v,
`
`
`
`-,'—IE'M.-z-"
`
`sir-€579armada”anaemia”:.,»-tor“"
`
`
`
`
`
`
`Maintenance Themp};
`
`Interferon After
`ConVentional Therapy
`The use of interferon alfa—2b (Intron
`A) for the maintenance of remission
`has been studied in detail in multiple
`randomized trials over the last 15 years.
`Some show no effect, some show a
`modest event—free survival benefit with-
`out an overall survival benefit, and a
`few demonstrate an overall survival ben-
`efit. The overall conclusion from this
`data remains controversial.
`Synthesizing these independent, con-
`
`Prognostic Factors[2,3]
`Disease Features
`
`- Clinical stage (hemoglobin, para—
`protein, calcium, renal iunction)[1 ,2)
`82—M[1]
`
`-
`
`- CFlP[1]
`
`-
`
`PCLJ[1]
`
`- Lactate dehydrogenasem
`
`- Presence of deletion 13 chromosome
`abnormality (for transplant)[47,71]
`
`- Mlcrovesseldensity[55,56]
`
`-
`
`-
`
`-
`
`Peripheral blood monoclonal plasma
`cells > 4%
`
`P—gp expression
`
`Soluble lL—S receptor
`
`- Serum (shed) CD56
`
`Plasmablastic morphology
`-
`Response Features
`
`- % of reduction: CFl, PR, vs SD or
`worse
`
`- Achievement of plateau
`
`- Duration of plateau[6]
`
`- Molecular CR vs clinical CH[1 6]
`——-———_
`
`B2—M = beta-2-microglobulin; CR = complete re—
`sponse; CFiP = C—reactive protein; |L-6 = interleu-
`kin-6: PCLI = plasma cell labeling index: P-gp =
`P-glycoprotein,‘ PFi = partial response; SD = sta-
`ble disease.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`flicting reports, a balanced conclu—
`sion[21,22] suggests that the effect of
`interferon therapy after a complete or
`partial response from conventional che—
`motherapy is, at best, a several—month
`improvement of event—free survival (but
`not overall survival) for a minority
`(< 15%) of patients. A 1998 meta-anal—
`ysis of 4,000 randomized patients, pre—
`sented in abstract form, concluded that
`the benefit is 7 months of overall sur—
`vival with P < .03.[23] Newly published
`studies with interferori randomization
`and various conventional treatments are
`similar to the earlier pattern—some—
`times with a significant event—free sur—
`vival
`advantage,
`but
`either
`a
`nOnsignificant overall survival advan—
`tage[24] or no advantage.[25,26]
`A decision to use interferon for pos—
`tremission maintenance should be made
`
`
`
`
`
`Hierarchy Of Responsesa
`
`- Progressive disease (> 25% increase
`in M protein or new bony lesion)
`
`- Stable disease/plateau phase
`
`- Minimal response
`
`PR (PR, > 50% decrease in M protein)
`-
`- Very good PFl (> 90% decrease in
`M protein)
`
`
`
`- CH (CH, undetectable paraprotein,
`low marrow plasma cell %)
`
`- CR (no clonal kappa/lambda
`population in marrow)
`
`- CR (no PCP—detectable clonal
`rearrangement in marrow)
`- Cure
`
`-—————.__
`
`ABMTH =Auto|ogous Blood and MarrowTransplant
`Registry; CR =complete response; EBMT: Euro~
`pean Blood and MarrowTransplant; lBMTFl = Inter—
`national Bcne MarrowTransplant Registries; PCR =
`polymerase chain reaction: PR = partial response.
`“The Annotation of the EBMT, IBMTR, and ABMTFl
`response definitions. authored by Blade et at, pro-
`vides a more detailed description, encompassing
`measurements otherthan just M protein.[72]
`
`recognizing that further study will be
`necessary to define which patient sub-
`sets may derive the most benefit.[22]
`Toxicity Ofinterferon at the typical dose
`Of 3 million units three times a week
`may include flu-like symptoms, depres—
`sion, and fatigue. For most, the expense,
`toxicity, and inconvenience of the in—
`jections will accrue no survival benefit.
`
`Steroids
`
`Steroids have also been applied for
`the purpose of postconventional thera—
`py maintenance. In the Southwest On—
`cology Group (SWOG) Study 9028,
`myeloma patients who had achieved at
`least a partial response after VAD
`chemotherapy were randomized be—
`tween interferon Or interferon/pred—
`nisone maintenance. The addition of
`prednisone to interferon resulted in a
`significant progression-free survival dif-
`ference (19 vs 9 months, P = .008), but
`a nonsignificant overall survival advan-
`tage (57 vs 46 months, P = .36).[27]
`
`Interferon After Transplant
`An increased frequency of very low
`
`tumor burden is associated with molec—
`
`75
`NCCN PROCEEDINGS - NOVEMBER 2000 - ONCOLOGY
`|PR2018—01714
`_ -‘,___Ce|eneo Ex. 2007, Page 5 ,L,
`
`‘
`
`IPR2018-01714
`Celgene Ex. 2007, Page 5
`
`

`

`
`
`ular complete response and an altered
`immunologic environment occurs fol—
`lowing high—dose chemotherapy with
`autologous stem—cell rescue. Conceptu-
`ally, this context may engender a more
`substantial clinical effect of interferon
`than in maintenance with conventional
`
`therapy.
`Phase II studies have shown the fea—
`
`sibility of posttranSplant interfer—
`on.[14, 15,28] A small phase III trial (42
`patients 011 each arm) showed improve-
`ment in event—free survival, but not
`survival benefit.[29] While the above-
`mentioned conclusion for interferOn
`maintenance after standard chemother—
`
`apy remissiou presumes a similar effect
`no matter which conventional chemo—
`
`therapy is used, it is possible that the
`conclusions ofphase III interferon-after—
`transplant trials may beregimen specific.
`
`Bisghosghonates
`
`Fracture Prevention
`
`including
`Bisphosphonate drugs,
`etidronate (Didronel), clodronate,
`alendronate (Fosamax), pamidronate
`(Aredia), and zoledronate (Zometa),
`bind tightly to the hydroxyapatite min-
`eral of bone, inhibiting osteoclast func-
`tion. This tips the balance of osteoclast
`activity from bone reabsorption to bone
`formation, with a net result of stronger
`bones. Additional mechanisms may in-
`clude induction of osteoclast apopto—
`sis[30] and possibly in vitro myeloma
`cell apoptosis.[31] Also, an immuno—
`logic effect of the bisphosphonates has
`been demonstrated:
`the induction of
`
`gamma—delta—restricted anti—myeloma T
`cells by aminobisphosphonates[32] re-
`mains to be clinically developed.
`Conventionally, the bisphosphonates
`are ranked by their antihypercalcemic
`potency. Zoledronate is more potent
`than pamidronate, which is more potent
`than clodronate or etidronate.[33] As
`an understanding develops about how
`to focus on mechanisms of action other
`than antihypercalcemic effects, it is con—
`ceivable that a primarily “antimyelo—
`ma” member of the drug class will be
`identified. Randomized phase III data
`are available for the application of etidr—
`onate,[34] clodronate,[35] and pami—
`dronate to the management of myel—
`oma.[36] For zoledronate, phase III
`trial data are not yet available.
`
`- Etidronate—In the earliest of these
`
`trials, myeloma patients received oral
`etidronate or placebo in conjunction
`with MP. Absence of a significant dif—
`ference in bone pain, episodes of hy-
`percalcemia, or development of
`pathologic fractures was observed.[34]
`In the clodroaate trial, myeloma pa—
`tients received oral clodronate or place-
`bo in conjunction with MP. The data
`showed that clodronate treatment re—
`
`sulted in a significantly decreased rate
`of progression of osteolytic lesions (12%
`vs 24%, P = .026), a nonsignificantly
`decreased rate of vertebral fractures
`
`(30% vs 40%, P not significant), and
`significantly better pain scores.[35]
`
`I Pamidronate—In the pamidronate
`phase III trial, originally published by
`Berenson et al in 1996 and updated in
`1998, patients receiving a variety of con—
`ventional chemotherapy regimens were
`randomized between monthly intrave—
`nous (IV) pamidronate vs placebo for
`21 cycles.[36] Considering the end point
`of “skeletal events," the pamidronate—
`treated group showed a statistically sig—
`nificant advantage over the placebo
`group (eg, 38% at2l cycles vs 51%, P:
`.015). There was no difference in the
`
`rates of hypercalcemia, chemotherapy
`response rates, and overall survival. A
`subset analysis, however, showed a me—
`dian survival benefit for patients receiv—
`ing pamidronate, when restricted to those
`patients who were receiving second-line
`chemotherapy (21 vs 14 months, P =
`.041, after adjustment). [36]
`
`- Clodronate—The initiation of bis—
`
`phosphonate treatment may occur at the
`onset of any bone lesion or osteopenia,
`or other treatment requiring chemother—
`apy; phase III data support its use.[37]
`The duration of treatment analyzed in
`the clodronate trial was daily for 24
`months;
`in the pamidronate study,
`it
`was up to 21 monthly IV cycles.
`Considering the good safety profile
`of the biphosphonates, an open-ended,
`individually tailored treatment duration
`can be recommended. The American
`
`Society of Clinical Oncology’ 5 (ASCO)
`1998 Guidelines on the Role of Bisphos-
`phonates in Breast Cancer have two
`points that may be applied: (1) once
`initiated, bisphosphonates may be con—
`tinued until there is evidence of a clini-
`cally assessed “substantial decline,” (2)
`their benefits consist of reduced skelee
`
`tal events and improved extent of pain,
`
`but not life prolongation.[38]
`Interval monitoring for skeletal pro—
`gression, with bone survey, is recom—
`mended at yearly intervals or for
`symptoms.[l] Patients on the pami—
`dronate treatment arm of the above—
`mentioned trial still had a 31 % incidence
`
`of ”any pathologic fracture” at 21
`months.[36] Similarly, interval moni—
`toring of calcium, renal function, and
`anemia may anticipate symptomatic
`presentation.
`
`Hypercalcemia
`Hydration, prednisone, and (posthy—
`dration) furosemide (Lasix) diuresis
`may be the initial treatment for hyper—
`calcemia. Bisphosphonate therapy—eg,
`90 mg IV pamidronate—will be fre—
`quently effective, and should be used if
`hypercalcemia has not resolved with
`less conservative measures.[39]
`
`Other Supportive Care
`
`Growth Factors
`Anemia, whether disease-related or
`
`treatment—related, is frequent in myelo—
`ma. Use of exogenous erythropoietin
`(Epogen, Procrit) therapy may improve
`both the hemoglobin level and sense of
`well—being.[39] Schedules,
`including
`erythropoietin 150 U/kg three times a
`week, 10,000 units three times a week,
`and 40,000 units once a week, have
`been advocated. The likelihood of re—
`
`sponse is influenced by the absence of
`other causes of anemia (vitamin Bu,
`folate, or iron deficiency), as well as
`the pretreatment (endogenous) erythro—
`poietin level. Increments of 2 g/dL of
`hemoglobin may be Seen in 60% to
`80% of myeloma patients.[40]
`
`
`
`Neutropenia
`Neutropenia may be related to marrow
`damage from chemotherapy, delayed
`recovery from high-dose therapy with
`stem-cell rescue, or disease progression.
`For chemotherapy-related neutropenia,
`the ASCO guidelines, developed main—
`ly for solid tumors, may be applied. [41]
`The prophylactic use of granulocyte
`colony—stimulating factor (G—CSF; Ne—
`upogen) or granulocyte-macr0phage
`colony—stimulating factor (GM—CSF;
`sargramostim, Leukine) should be lim—
`ited to those patients who have had hos—
`pitalization for neutropenia—related
`infection, but not for uncomplicated fe-
`brile (or afebrile) neutropenia. For neu-
`
`76
`
`ONCOLOGY - VOLUME 14 - NUMBER 11A ' NCCN PROCEEDINGS
`
`|PR2018—01714
`Celgene Ex. 2007, Page 6
`
`_
`‘7.—
`
`IPR2018-01714
`Celgene Ex. 2007, Page 6
`
`

`

`tropenia related to disease progression,
`treatment of the myeloma should be the
`priority. Growth factors may also be
`used as part of a stem—cell mobilization
`regimen.
`
`Infection
`
`Antibiotic prophylaxis, with cotri—
`maxozole (trimethoprim/sulfamethox-
`azole; Bactrim, Septra, Sulfatrim) or
`penicillin, in the first months of chemo-
`therapy (or in the context of stem-cell
`transplant) may be useful. Immuniza—
`tion for prevention of pneumococcal or
`influenza infection is also recommend
`
`ed—notwithstanding a likely subopti—
`mal antibody response.[39]
`
`Renal Failure
`
`An elevated creatinine level in my—
`eloma may be a consequence of hyper—
`caICemia, disease-related kidney
`complications (in which light chains of
`the paraprotein accumulate in the col—
`lecting tubules), amyloid deposition
`(with more prominent albuminuria), or
`other intercurrent renal disease. Owing
`to its rapid response, VAD is a regimen
`that may be useful for resolution of
`renal failure.[39] The successful appli—
`cation of high-dose chemotherapy in
`patients with renal failure has been
`reported. [42]
`
`High-Dose Chemotherapy
`
`Since the first phase II series was
`reported in 1983,[43] a variety of regiu
`mens, predominantly based on high—
`dose melphalan, uSually 140 mgi’m2
`(MELl40), have been published. Some
`feasible transplant choices are summa—
`rized in Table 7. Patients treated on
`
`these protocols represent a relatively
`large phase II experience.
`To fairly assess the benefit of high—
`dose chemotherapy, it is useful to quan—
`titate the heterogeneity of prognosis,
`even among conventionally treated pa—
`tients. Blade et a1[9] and Oivanen[6]
`published reports that identify a good—
`prognosis subset among patients treat-
`ed without high-dose therapy. Blade et
`a1 analyzed the results of 487 myeloma
`patients treated with conventional che—
`motherapy in trials between 1985 and
`1989.
`
`This analysis shows that survival
`among the 77 patients who appeared to
`be eligible for bone marrow transplant
`(age < 65; stage H or III; performance
`
`status = 0, l, 2; chemotherapy~respon~
`sive disease) was 60 months—much
`better than the overall median of 29
`months.[9] Similarly, Oivanen’s analy—
`sis of conventionally treated myeloma
`patients in Finnish trials shows a subset
`(those with the longest initial plateau
`response) of patients with a median sur—
`vival of 81 months compared to 44
`months for the entire group.[6]
`The magnitude of survival differenc—
`es shows that it may be unclear what
`standard results of single—arm, phase II
`trials of high-dose chemotherapy in
`patients with myeloma should be com~
`pared against. Some of the analyses
`described below use a pair—matching
`system to estimate what the prognosis
`of the high—dose chemotherapy group
`would have been if treated convention—
`ally. Phase III data are described
`separately.
`An alternative perspective regard—
`ing the overall effect of high-dose che—
`motherapy is found in the analysis by
`the Nordic Myeloma Study Group. This
`report analyzed outcomes of (nonran—
`domized) trial-registered myeloma
`cases, which accounted for about three—
`quarters of the predicted total number
`of cases in Denmark, Norway, and Swe-
`den. Their population—based analysis
`shows that the introduction of the trans—
`
`plant technique is associated with a sur—
`vival advantage, whether the apparently
`tranSplant—eligible patients (relative risk
`[RR] = 1.62; 95% confidence interval
`[CI] = 1.22—2.15; P = .001) or the entire
`trial-registered population (RR = 1.46;
`95% CI = 1.14—1.86; P = .002) were
`considered.[44] This analysis did not
`consider early vs late transplant;
`the
`214 transplants were all within 1 year
`of the start of VAD therapy.
`
`Phase II Data
`
`High-Dose Chemotherapy
`The SWOG 8993 trial enrolled pa—
`tients with myeloma < 70 years old
`with progressive or refractory myelo—
`ma, despite alkylating agent, dexa—
`methasone, or VAD treatment.[20] The
`high-dose chemotherapy consisted of
`cyclophosphamide 6 g/mz, followed by
`apheresis of stem cells. Patients (and
`the autograft) were then rescreened for
`eligibility, followed by melphalan 100
`mg/m2 on days —4 and —3 (MELZOO),
`and infusion of