`
`NUMBER 5
`
`- FEBRUARY ‘10 2010
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`ORIGINAL REPORT
`
`Phase I/II Trial of Metronomic Chemotherapy With Daily
`Dalteparin and Cyclophosphamide, Twice-Weekly
`Methotrexate, and Daily Prednisone As Therapy for
`Metastatic Breast Cancer Using Vascular Endothelial
`Growth Factor and Soluble Vascular Endothelial Growth
`
`Factor Receptor Levels As Markers of Response
`Nan Soon Wong, Robert A. Buckman, Mark Clemons, Shailendra Verma, Susan Dent, Maureen E. Trudeau,
`Kathie Roche, John Ebos, Robert Kerbel, Gerrit E. DeBoer, Donald ].A. Sutherland, Urban Emmenegger,
`Joyce Slingerland, Sandra Gardner, and Kathleen I. Pritchard
`
`ABSTRACT
`
`Purpose
`Preclinical studies indicate that metronomic chemotherapy is antiangiogenic and synergistic with
`other antiangiogenic agents. We designed a phase l/ll study to evaluate the safety and activity of
`adding dalteparin and prednisone to metronomic cyclophosphamide and methotrexate in women
`with measurable metastatic breast cancer (MBC).
`Patients and Methods
`Patients received daily dalteparin and oral cyclophosphamide, twice—weekly methotrexate, and
`daily prednisone (daICMP). The primary study end point was clinical benefit rate (CBR), a
`combination of complete response (CR), partial response (PR), and prolonged stable disease for
`2 24 weeks (pSD). Secondary end points included time to progression (TTP), duration of response,
`and overall survival (OS). Biomarker response to treatment was assessed by using plasma vascular
`endothelial growth factor (VEGF) and soluble VEGF receptors (sVEGFRs) —1 and —2.
`Results
`Forty-one eligible patients were accrued. Sixteen (39%) had no prior chemotherapy for MBC; 15
`(37%) had two or more chemotherapy regimens for MBC. Toxicities were minimal except for
`transient grade 3 elevation of liver transaminases in 11 patients (27%) and grade 3 vomiting in one
`patient (2%). One patient (2%) had CR, six (15%) had PR, and three (7%) had pSD, for a CBR of
`10 (24%) of 41 patients. Median 'I'I'P was 10 weeks (95% Cl, 8 to 17 weeks), and median OS was
`48 weeks (95% CI, 32 to 79 weeks). VEGF levels decreased but not significantly, whereas
`sVEGFR-i and -2 levels increased significantly after 2 weeks of therapy. There was no correlation
`between response and VEGF, sVEGFR-i, or sVEGFR-2 levels.
`Conclusion
`Metronomic daICMP is safe, well tolerated, and clinically active in MBC.
`
`J Clin Oncol 28:723-730. (6) 2009 by American Society of Clinical Oncology
`
`From the Sunnybrook Odette Cancer
`Centre, University of Toronto, Toronto;
`Princess Margaret Hospital; Ottawa
`Hospital Cancer Centre, University of
`Ottawa; and Ontario Clinical Oncology
`Group, Ontario, Canada.
`Submitted May 26, 2009; accepted
`October 2, 2009; published online
`ahead of print at www.jco.org on
`December 21, 2009.
`
`Supported in part by a grant from Pfizer
`(formerly Pharmacia Upjohn).
`
`Authors’ disclosures of potential con—
`flicts of interest and author contribu-
`tions are found at the end of this
`article.
`
`Corresponding author: Kathleen l.
`Pritchard, MD, FRCP, Division of Medi-
`cal Oncology, Sunnybrook Odette
`Cancer Centre, Ontario Clinical Oncol—
`ogy Group, University of Toronto, 2075
`Bayview Ave, Toronto M4N3M5,
`Ontario. Canada; e-mail: kathy
`.pritchard@sunnybrook.ca.
`© 2009 by American Society of Clinical
`Oncology
`0732-1 83X/10/2805—723/$20t00
`DOI: 10.1200/JCO.200924.0143
`
`lNTRODUCTION
`
`Metronomic chemotherapy refers to treatment at
`regular, close intervals without prolonged breaks at
`doses significantly less than the maximum-tolerated
`close.1 The theoretical basis for metronomic chem—
`
`induce endogenous inhibitors of angiogenesis,
`such as thrombospondin-l.S Furthermore, low
`concentrations of drug retain activity against endo—
`thelial cells.”
`Preclinical studies demonstrate that metro-
`
`nomic chemotherapy inhibits tumor growth, can
`control drug resistant tumors more effectively
`otherapy is that conventional cytotoxic drugs target
`than conventional schedules,6 exhibits synergism
`vascular endothelial cell proliferation“6 but do not
`achieve sustained antiangiogenesis because of en—
`when combined with antiangiogenic agentsf‘8
`dothelial cell recovery during treatment breaks,
`and controls tumor growth through inhibition of
`which may be overcome by frequent treatment at
`endothelial cellsg’lo and circulating endothelial
`low doses. Metronomic chemotherapy also may
`progenitor cells.‘1
`Genentech 2089 - Celltrion v. Genentech - |PR2017-01122
`© 2009 by American Society of Clinical Oncology
`
`723
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`CoDVright © 2017 American Society of Clinical Oncolorzv. All rights reserved.
`
`Genentech 2089 - Celltrion v. Genentech - IPR2017-01122
`
`
`
`Wong et al
`
`
`
`The first trial to assess the efficacy ofmetronomic chemotherapy
`for metastatic breast cancer was reported by Colleoni et al.12 In this
`phase II study using daily oral cyclophosphamide and twice—weekly
`methotrexate, the overall response rate (ORR) was 19% (95% CI,
`10.2% to 30.9%) with minimal toxicity. Plasma vascular endothelial
`growth factor (VEGF) levels decreased in all patients on therapy for at
`least 2 months.
`
`On the basis ofpreclinical evidence ofsynergism between metro—
`nomic chemotherapy and antiangiogenic agents, and on the basis of
`the postulated antiangiogenic properties of corticosteroids”15 and
`low—molecular—weight heparins,16 we designed a phase I/II trial com-
`bining daily dalteparin, cyclophosphamide, twice-weekly methotrex-
`ate, and daily prednisone (dalCMP) as therapy for MBC. This
`concept17 and preliminary data concerning these trial results18 were
`presented previously. Doses of cyclophosphamide and methotrexate
`were adopted from Colleoni et al,12 whereas dalteparin and pred-
`nisone were given in low doses that were thought to be minimally toxic
`but potentially antiangiogenic.
`Elevated plasma VEGF correlates with advanced disease stage
`and progression.19'21Changes in VEGF level correlate with treatment
`response in several studies.”25 Plasma VEGF has been a proposed
`biomarker in determining the optimal dose of antiangiogenic
`drugs.26 Soluble VEGF receptor (sVEGFR) —1 is an endogenous
`inhibitor of VEGF that acts through sequestration of VEGF.27 The
`presence of naturally occurring sVEGFR—2 in mouse and human
`plasma also has been reported.28 Both sVEGFR—l and sVEGFR—Z have
`been studied as possible biomarkers of response to antiangionenic
`therapy?”30 Therefore, we also assessed whether plasma VEGF or
`sVEGFR—l and -2 could serve as markers of antiangiogenic activity or
`clinical response.
`
`PATIENTS AND METHODS
`
`Eligibility
`Eligible patients met the following criteria: histologically confirmed
`breast cancer with metastatic disease; unidirnensionally measurable disease;
`Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2;
`life expectancy Z 6 months; white blood count 2 3.0 X 109/1.- neutrophil
`count 2 1.5 X log/L; platelet count 2 100 X 109/1,- creatinine S 177 umol/I;
`bilirubin S 1.5 times the upper limit of normal (ULN); and ALT, AST and
`alkaline phosphatase 5 2.5 X ULN. All patients had to have been considered
`for standard therapy for metastatic disease. Most patients were treated with
`dalCMP after standard anthracycline and taxane therapy for metastatic disease
`or as first—line therapy for metastatic disease if less toxic chemotherapy was
`appropriate. Signed informed consent was obtained from all patients. The
`study protocol received approval from the research ethics board of each par—
`ticipating hospital (CONSORT diagram, Fig 1).
`
`Investigations and Treatment
`Baseline evaluation included history and physical examination, com—
`plete blood count, differential count, creatinine, liver function tests (ie, AST,
`ALT, alkaline phosphatase), prothrombin time, partial thromboplasfin time,
`plasma VEGF, sVEGFR—l, sVEGFR—Z and chest x—ray. Tumor assessment was
`performed by using computed tomography scan, magnetic resonance imag-
`ing, or ultrasound at baseline and then every 8 weeks. All patients were to
`receive, within 14 days of study entry, cyclophosphamide 50 mg orally daily,
`methotrexate 2.5 mg twice daily by mouth twice weekly, dalteparin 5,000 U
`subcutaneously daily, and prednisone 5 mg by orally daily in continuous 28
`day cycles until disease progression or unacceptable toxicity occurred. Toxicity
`was graded by National Cancer Institute Common Toxicity Criteria (version
`
`
`
`Assessed for Eligibility
`(N = 44)
`
`Not meeting inclusion criteria in = 3)
`
`Allocated to intervention
`
`Received allocated intervention
`
`(n = 41)
`
`(n = 40)
`
`Did not receive allocated intervention
`
`in = 1)
`
`
`
`Lost to follow-up
`
`Discontinued inten/ention
`
`Analyzed for efficacy
`
`Analyzed for safety
`
`Excluded from efficacy analysis
`
`in : 0)
`
`(n = O)
`
`(n = 41)
`
`(n = 40)
`
`in = 0)
`
`in = 1)
`Excluded from safety analysis
`
`
`Fig 1. CONSORT diagram.
`
`2). Dose levels and adjustments on the basis oftoxicity were used in all patients
`(Tables 1 and 2). Re—escalation of doses was attempted if toxicity resolved.
`Toxicity assessments, complete blood count, differential count, cre-
`atinine and liver function tests were performed weekly for 4 weeks and then
`monthly. Clinical tumor assessments were completed monthly. Radiologic
`tumor assessments were performed every 8 weeks. Plasma for VEGF,
`sVEGFR—l, and sVEGFR-Z analyses was collected at 2 weeks, 8 weeks, and 24
`weeks after beginning therapy. Plasma measurements for VEGF, sVEGF—l,
`and sVEGF-2 were done after 2 weeks to correlate this early result with
`subsequent tumor response.
`
`Study End Points
`The primary outcome measure was clinical benefit rate (CBR), which
`was defined as complete response (CR), partial response (PR), or prolonged
`stable disease 2 24 weeks (pSD). Response was evaluated by Response Evalu-
`ation Criteria in Solid Tumors (RECIST). Time to progression (TTP) was
`calculated from first treatment date until date of first documentation of pro-
`gressive disease (PD) or related death ifthis occurred before PD. Patients with
`death unrelated to breast cancer or dalCMP were censored in 'I'TP analysis.
`Overall survival (05) was calculated from first treatment date until date of
`death as a result ofany cause. Patients without an event and alive at data cutoff
`were censored in OS analysis. Duration of response (DOR) was calculated
`from date offirst documented objective response (ie, CR or PR) or date offirst
`
`Table 1. Treatment Dose Levels
`
`
`
`Drug
`Cyclophosphamide
`
`Methotrexate
`
`-—2
`25 mg every
`other day
`2.5 mg Monday
`
`Dalteparin
`Prednisone
`
`5.000 U daily
`5 mg daily
`
`Dose Level
`
`—1
`25 mg daily
`
`2.5 mg Monday,
`Tuesday
`5,000 U daily
`5 mg daily
`
`Starting Dose
`50 mg daily
`
`5 mg Monday,
`Tuesday
`5,000 U daily
`5 mg daily
`
`724
`
`© 2009 by American Society of Clinical Oncology
`
`JOURNAL or CLINICAL ONCOLOGY
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`CODVrlEl‘lt © 2017 American Society of Clinical Oncololzv. All rights reserved.
`
`
`
`
`
`Metronomic DalCMP for Metastatic Breast Cancer
`
`Table 2. Dose Modification for Hematologic and GI Toxicity
`
`Toxicity
`Grade
`0-1
`2
`
`Neutrophil
`
`Full dose, no delay
`Delay cyclophosphamide and methotrexate until
`grade 1 toxicity or less; resume at full dose.
`If delay longer than 2 weeks, resume at one
`dose level reduction‘
`Delay cyclophosphamide and methotrexate until
`grade 1 toxicity or less; resume at one dose
`level reduction"
`
`Toxicity
`Platelet
`
`ALT/AST
`
`Full dose, no delay
`Delay cyclophosphamide and methotrexate until
`grade 1 toxicity or less; resume at full dose
`
`Full dose, no delay
`Full dose, no delay
`
`Delay cyclophosphamide. methotrexate, and
`dalteparin until grade 1 toxicity or less;
`resume at one dose level reduction‘
`
`Reduce methotrexate one dose level'
`
`
`
`*If no additional toxicity was observed, doses were increased one level per course until starting dose was reached.
`
`therapy (ie, pSD) until date of first documented PD. For DOR analysis, pa-
`tients were censored ifthey were alive without progressionor ifthey were dead
`as a result of unrelated causes at data cutoff.
`
`VEGF, sVEFGR- 1, and sVEGFR-Z Protein Measurement
`Commercially available human sandwich enzyme—linked immunosor-
`bant assays (ELISA; R&D Systems, Minneapolis, MN) to measure VEGF
`(catalogue # DVEOO), sVEGFR—Z (catalogue # DVRZOO), and sVEGFR—l (cat-
`alogue # DVR100) were used according to manufacturer instructions. All
`ELISA kits were composed ofa monoclonal capture antibody and horseradish
`peroxidase— conjugated polyclonal antibody used for detection with limits of
`no less than 31.2, 78.1, and 31.2 pg/mL, respectively. All samples were mea-
`sured in duplicate, and absorbance was determined by using the microplate
`reader Benchmark Plus (Bio-Rad Laboratories, Philadelphia, PA) set to 450
`nm with wavelength correction set to 540 nm.
`Statistical Methods
`
`For this nonrandomized, phase I/II study, sample size calculation was
`based on an optimal two-stage design with 90% power and an a of .05.3 1 A
`CBR of 20% was chosen as relevant (ie, alternate hypothesis, H1), whereas
`S 5% indicated insufficient activity (ie, null hypothesis, Ho). Twenty—one
`patients were required for the first stage, and early termination occurred if
`CBR was observed in fewer than two patients. Twenty additional patients
`were recruited to the second stage if CBR was observed in two or more
`patients, which provided a total sample size of 41 patients. The null
`hypothesis was rejected if CBR was observed in five or more patients in
`both stages combined. 95% CIs for proportions were derived by using the
`exact binomial method.32 Response rate comparisons were made by using
`Fisher’s exact test.33 Survival estimates were derived by the Kaplan and
`Meier method.34 Biomarker values were summarized at baseline and after
`2, 8, and 24 weeks by using medians and distribution-free 95% CIs for the
`median. Relative differences in biomarkers between baseline and other
`
`time points were calculated for each patient. The Wilcoxon signed rank
`test35 was used to assess whether median relative marker changes were
`significantly different from baseline, and the Wilcoxon rank sum test35 was
`used to compare those with CBR to those with PD at baseline and for
`relative differences across time. Analysis was performed by using SAS
`version 9.1.3 (SAS Institute, Cary, NC).
`
`
`
`Patient Characteristics
`
`Between October 2003 and February 2005, 44 patients were re-
`cruited from two cancer centers in Ontario, Canada. Three patients
`were ineligible (n = 1 for no measurable disease, 11 = 1 for bilirubin
`greater than 2 X ULN, and n = 1 for lyrnphangitis carcinomatosis
`with a life expectancy S 6 months). Table 3 lists baseline characteris-
`tics of the 41 eligible patients. Thirty (73%) had visceral disease.
`
`Sixteen (39%) never received previous chemotherapy for MBC,
`whereas 15 (37%) had received two or more regimens of chemother-
`apy for MBC. Eighteen (44%) were refractory to anthracyclines and
`taxanes and had either experienced relapse within 1 year of receiving
`these as adjuvant therapy, had received both agents for MBC, or had
`received an adjuvant anthracycline followed by a taxane for MBC.
`
`Treatment Delivery
`One hundred sixty treatment cycles were delivered (median,
`three; range, zero to 15). Five patients required per-protocol dose
`reduction to — 1 level (n = 4 for grade 3 elevation oftransaminases and
`n = l for grade 2 elevation oftransaminases). Three required —2 level
`dose reduction (for grade 3 elevation of transaminases). Dose inter-
`ruption occurred in seven patients; for 1 day in two patients (n = 1 for
`hemoptysis with known lung metastasis, n = 1 for chest infection); for
`1 week in two patients (n = l for grade 3 elevation of transaminases,
`n = 1 for pyrexia of unknown origin); for 2 weeks in one patient (for
`decline in performance status); and for 3 weeks in two patients (n = l
`for hydronephrosis, n = 1 for dyspepsia).
`
`Toxicity
`Forty patients were evaluable for toxicity; one patient was eligible
`for study but did not receive any treatment. Selected toxicities are
`listed in Table 4. N0 grade 3 or 4 hematologic toxicity was observed.
`One patient (2%) had grade 3 fatigue present at baseline. One patient
`(2%) developed transient, grade 1 hemoptysis that resolved without
`treatment. Five patients (12.5%) developed grade 2 infection without
`associated neutropenia (n = 1 with urinary tract infection, 11 = 1 with
`pneumonia, n = 1 with pyrexia of unknown origin, 11 = 1 with upper
`respiratory tract infection, and n = 1 with upper limb cellulitis related
`to lyrnphedema), and these patients received outpatient oral antibiot—
`ics. Eleven patients (27.5%) developed grade 3 elevation of transami-
`nases. Of these, four had no documented liver metastasis, two had
`
`stable liver metastases, and five had progressive liver metastases.
`Among patients with progressive liver metastases, transaminases im—
`proved with cessation of methotrexate in two and remained persis—
`tently elevated in one, whereas no additional follow-up transaminases
`were available in two patients. Among patients without progressive
`liver metastases, transaminases improved in all six with reduction
`of dose (11 = 5) or temporary (ie, 3 weeks) cessation of methotrex—
`ate (n = 1).
`
`www.jco.org
`
`© 2009 by American Society of Clinical Oncology
`
`725
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`
`
`Wong et al
`
`Table 3. Demographic and Baseline Clinical Characteristics
`Patients
`(N = 41)
`
`Characteristic
`
`No.
`
`%
`
`Age, years
`Median
`Range
`ECOG performance status
`
`
`
`Hormone receptor status
`ER or PR positive
`ER and PR negative
`Unknown
`HEFlZ/neu status
`Positive
`Negative
`Unknown
`Sites of metastases
`Visceral dominant
`Nonvisceral
`Three or more sites
`Prior disease-free survival, years
`a 1
`< 1
`Metastatic at presentation
`Prior adjuvant endocrine therapy
`Prior adjuvant chemotherapy
`Prior lines of endocrine therapy for advanced disease
`0/not applicable
`1
`2
`> 2
`Prior lines of chemotherapy for advanced disease
`
`> 2
`All prior endocrine therapy lines‘
`0
`1
`2
`> 2
`All prior chemotherapy lines'
`
`> 2
`
`2.5
`
`19.5
`58.5
`22.0
`
`73.2
`26.8
`26.8
`
`9.8
`82.9
`73
`41.5
`68.3
`
`34.1
`9.8
`24.4
`31.7
`
`39.0
`24.4
`12.2
`24.4
`
`24.4
`9.8
`17.1
`48.8
`
`12.2
`22.0
`. 26.8
`39.0
`
`11
`16
`
`among patients with CR or PR was 22 weeks (range, 12 to Z 136
`weeks). Median DOR for pSD was 37 weeks (range, 36 to 39 weeks).
`Median TTP was 10 weeks (95% CI, 7.7 to 17.3 weeks). Median OS
`was 48 weeks (95% CI, 32 to 79 weeks; Figs 2A, 2B, and 2C).
`Among 16 patients with no prior chemotherapy for MBC, one
`(6%) had CR, four (25%) had PR, and none had pSD, for an OR of
`five (31%) of 16 patients and a CBR offive (31%) of 16 patients. Three
`patients (19%) had SD, and eight patients (50%) had PD. Among 15
`patients who had two or more prior regimens of chemotherapy for
`MBC, one had PR and three had pSD, for an OR of7% and a CBR of
`27% (four of 15 patients), whereas 11 patients (73%) had PD. When
`patients were categorized into those who had no prior chemotherapy
`for MBC, those who had received one or two prior chemotherapy
`regimens for MBC, and those who had greater than two prior regi—
`mens, there was no significant difference in OR (P = .22) or CBR
`(P = .55) across the categories. We also found no significant relation-
`ship between response and disease—free interval (DFI; < 1 year v 2 1
`year) hormone receptor status (estrogen receptor positive with or
`without progesterone receptor positive v estrogen and progester-
`one receptors negative), or HERZ status (HERZ positive, negative,
`or unknown).
`
`Biomarker Results
`
`Serial assessment of plasma biomarkers was available for 33 pa-
`tients at baseline, for 27 patients after 2 weeks, for 23 patients after 8
`weeks, and for nine patients after 24 weeks. After 2 weeks, there was a
`median 56% relative decrease in serum VEGF levels (95% CI, —67%
`to 0%; P = .016), a median 17% relative increase in sVEGFR—l levels
`(95% CI, —04% to 83%; P = .01), and a median 7% relative increase
`in sVEGFR-Z levels (95% CI, 4% to 13%; P = .0002) from baseline.
`After 8 weeks, there was a median 64% relative decrease in VEGF levels
`(95% CI, -84% to 7%; P = .10), a median 21% relative increase in
`sVEGFR—l levels (95% CI, — 11% to 150%; P = .03), and a median 1%
`relative increase in sVEGFR—2 levels (95% CI, —5% to 14%; P = .46)
`from baseline. After 24 weeks, there was a median 43% relative de-
`crease in VEGF levels (95% CI, — 86% to 46%; P = .43), a median 37%
`relative increase in sVEGFR—l
`levels (95% CI, — 14% to 324%;
`P = .10), and a median 1% relative decrease in sVEGFR—Z levels (95%
`CI, — 1% to 13%; P = 1.00) from baseline. The median relative
`differences for all three biomarkers were similar for responders and
`nonresponders at baseline and after weeks 2, 8 and 24.
`
`In this study of metronomic dalCMP for treatment of MBC, we
`documented OR in seven patients (17%) and pSD in three patients
`(7%) to provide a CBR of 24%. We included pSD in the definition of
`clinical benefit because ofthe anticipated slow onset of response with
`metronomic chemotherapy and because of the recognition that ab-
`sence of progression is a reasonable therapeutic end point for
`women with MBC.
`
`In an unplanned post hoc analysis, we found OR in five (31%)
`of 16 patients who had no prior chemotherapy for MBC, which
`suggests that dalCMP could be additionally explored as first—line ther—
`apy, especially in patients without life-threatening disease. This strat-
`egy has the advantage of palliation without incurring the toxicities
`associated with the current standard first-line chemotherapy. Al-
`though there was only one overall response (7%) in 15 patients who
`
`JOURNAL OF CLINICAL ONCOUOGY
`
`ECOG, Eastern Cooperative Oncology Group; ER, estrogen receptor; PR,
`progresterone receptor.
`‘Refers to all lines of therapy,
`advanced disease.
`
`including adjuvant therapy and therapy for
`
`Efficacy
`One patient (2%) had CR, six (15%) had PR, and three (7%) had
`pSD, which provided an OR ofseven (17%) of41 patients (95% CI,
`7% to 32%) and a CBR of 10 (24%) of 41 patients (95% CI, 12% to
`40%). Seven patients ( 17%) had stable disease (SD) for greater than 12
`weeks but not for 24 weeks or longer, whereas 24 patients (59%) had
`PD. Responder characteristics are listed in Table 5. Median DOR
`
`726
`
`© 2009 by American Society of Clinical Oncology
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`Copyright © 2017 American Society of Clinical Oncologv. All rights reserved.
`
`
`
`Metronomic DalCMP for Metastatic Breast Cancer
`
`
`
`Toxicity
`Hematologic
`Anemia
`Leukopenia
`Neutropenia
`Thrombocytopenia
`Nonhematologic
`Alopecia
`Bleeding
`Diarrhea
`Fatigue
`Nausea/vomiting
`Sepsis
`Stomatitis
`Transaminitis
`
`No.
`
`30
`23
`39
`34
`
`39
`39
`32
`15
`26
`35
`40
`3
`
`NOTE. No. of patients = 40,
`Abbreviation: NA, not applicable.
`
`Table 4. Worst Hematologic and Nonhematologic Toxicity
`
`0
`
`1
`
`Toxicity by Grade
`2
`
`3
`
`4
`
`%
`
`75.0
`57.5
`97.5
`85.0
`
`97.5
`97.5
`80,0
`37.5
`65.0
`87.5
`100
`7.5
`
`No.
`
`
`%
`No.
`%
`
`No.
`
`%
`
`No.
`
`%
`
`6
`14
`1
`6
`
`1
`1
`8
`17
`9
`0
`0
`IO
`
`15.0
`35.0
`2.5
`1 5.0
`
`2.5
`2.5
`20.0
`42.5
`22.5
`
`25.0
`
`4
`3
`0
`0
`
`0
`0
`0
`7
`4
`5
`0
`16
`
`10.0
`7.5
`
`17.5
`10.0
`12.5
`
`40
`
`0
`0
`0
`O
`
`NA
`0
`O
`1
`1
`0
`0
`11
`
`2.5
`2.5
`
`27.5
`
`O
`O
`O
`0
`
`NA
`0
`0
`0
`0
`O
`O
`0
`
`
`
`had prior regimens of chemotherapy, pSD was observed in three
`patients (20%) to provide a CBR of 27%. This encouraging result
`suggests that additional comparison with other third- or fourth-line
`regimens also is warranted. We are currently expanding this phase I/II
`study to include more patients either with anthracycline—na'ive and
`taxane-na‘ive disease or with refractory disease to better assess re-
`sponse rates in these two groups of patients.
`Therapy was well tolerated; no grade 3 or 4 hematologic toxicity
`or sepsis and no grade 2 alopecia occurred. One patient developed
`grade 3 vomiting, which was attributed to uremia from obstructive
`nephropathy. The main toxicity was reversible grade 3 elevation of
`liver transaminases, which was easily managed with dose reduction
`with or without temporary dose interruption. Monthly monitoring of
`transaminases is recommended with this regimen. dalCMP required
`daily subcutaneous dalteparin, but compliance was high. Only one
`patient declined treatment after study commencement.
`
`We found no correlation between treatment response and VEGF
`or sVEGFR-l or —2 baseline levels or changes after 2, 8 or 24 weeks of
`therapy. Future studies may incorporate dynamic contrast-enhanced
`magnetic resonance imaging and measurement of circulating endo—
`thelial cells (CECs) and endothelial progenitor cells with more useful
`results.36 Similarly, the optimal biologic or antiangiogenic doses of
`cyclophosphamide and methotrexate have not been determined clin—
`ically37 and were derived from previous phase II data.12 Furthermore,
`the relative contribution of dalteparin and prednisone to the regimen
`could not be assessed. A useful study would be one that assesses
`changes in circulating endothelial progenitor levels with sequential
`addition ofeach component ofmetronomic chemotherapy over time.
`Two other recently published phase II trials”’39 explored metro-
`nomic chemotherapy together with the antiangiogenic agent bevaci-
`zumab. One single-arm, phase II trial in MBC38 reported an objective
`response rate of46% and a pSD rate of41 % for daily oral capecitabine
`
`Patient
`
`Response
`
`Length of Response
`(weeks)
`
`ECOG
`
`TS-1
`TS-2
`TS-3
`TS-7
`TS-10
`TS-1 1
`TS-25
`TS-34
`TS-35
`TS-37
`
`pSD
`Partial
`pSD
`Partial
`Partial
`CR
`Partial
`Partial
`pSD
`Partial
`
`39
`1 2
`37
`1 2
`22
`61
`
`1 6
`36
`23
`
`ON—‘OOOON—‘—I
`
`Table 5. Clinical Characteristics of Responders
`Characteristic
`
`Dominant Metastatic
`Sites
`Visceral
`Visceral
`Nonvisceral
`Nonvisceral
`Nonvisceral
`Visceral
`Nonvisceral
`Visceral
`Nonvisceral
`Nonvisceral
`
`No, of Involved
`Sites
`‘-
`
`__\NN_\_\_\_IN
`
`Prior DFl
`(years)
`> 1
`> 1
`> 1
`> 1
`< 1
`> 1
`> 1
`Metastatic'
`> 1
`.
`Metastatic'
`
`Prior Lines of
`Chemotherapy
`N
`
`dmNOOOOhO
`
`ER/PR
`Status
`+
`+
`+
`
`HEFIZ Status
`_
`Unknown
`Unknown
`
`
`
`Abbreviations: ECOG, Eastern Cooperative Oncology Group; DFI, disease-free interval; ER, estrogen receptor; PR, progesterone receptor; pSD. prolonged stable
`disease; CR, complete response.
`"Metastatic at presentation.
`
`www.jco. org
`
`© 2009 by American Society of Clinical Oncology
`
`727
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`Convrieht © 2017 American Society of Clinical Oncolozv. All rights reserved.
`
`
`
`>
`
`Wong et al
`
`
`
`
`
`Progression-FreeSurvival(%) 9
`
`No. at risk 41
`
`30
`
`40
`
`50
`
`Time (weeks)
`7
`1
`
`7 F
`
`A randomized, phase II trial of neoadjuvant letrozole versus
`letrozole plus low-dose daily oral cyclophosphamide in 114 elderly
`women with T2-4, NO-l, ER-positive breast cancer suggested good
`activity (OR, 87.7%) for the combination, whereas letrozole alone
`produced an ORR of71 .9% (not significantly different). Response was
`significantly correlated with postoperative reduction in levels ofKi67
`and VEGF A expression in tumor tissue.40
`It is also interesting to observe that, among three recently pub—
`lished randomized trials that compared taxane regimens with or with-
`out the antiangiogenic agent bevacizumab, the biggest additive effect
`of bevacizumab was seen in the study that used more metronomic
`dosing. Thus, the ECOG 2100 study,41 which compared weekly pacli—
`taxel with or without bevacizumab showed a significantly greater
`improvement in TTP than did either the AVADO42 or RIBBON—143
`study, each ofwhich studied every—3 —weeks docetaxel with or without
`bevacizumab (Avastin; Genentech Inc, South San Francisco, CA). It
`may be that weekly metronomic chemotherapy in addition to the
`antiangiogenic bevacizumab is a better schedule to explore.
`Burstein et al44 presented results from a randomized, phase II
`study of 55 women with MBC who were treated with metronomic
`chemotherapy in the first- or second—line setting. The ORR was 10%
`(95% CI,1% to 30%) among patients who received metronomic cy—
`clophosphamide and methotrexate compared with 29% (95% CI,
`15% to 50%) for those who received metronomic chemotherapy plus
`bevacizumab; these results led to early termination ofthe former arm
`even though response rates were not statistically significantly different.
`These findings and the results of this phase I/II study support the
`preclinical observation ofsynergism between metronomic chemother—
`apy and other antiangiogenic agents. In contrast, another recently
`reported, randomized, phase II study allocated 178 patients with MBC
`to receive either oral methotrexate and cyclophosphamide or the same
`regimen plus thalidomide.45 The overall clinical benefit was 47% for
`both arms, but more neurologic toxicity (grade 2 2) and a higher
`incidence ofthromboembolic events occurred in patients allocated to
`the thalidomide arm.45 These discordant findings underscore the need
`for future studies that use appropriate biomarkers to select the optimal
`agent(s), schedule, and biologic dose of metronomic chemotherapy
`for combination with the appropriate antiangiogenic drug and for the
`use of randomized designs whenever feasible.
`In conclusion, metronomic dalCMP is a safe, well-tolerated, and
`active regimen in the treatment of MBC. This regimen can be used in
`patients who want effective, but relatively nontoxic, therapy in the early
`metastatic setting or as third-, fourth— or fifth-line therapy in those who
`are resistant to more—standard chemotherapy for metastatic disease.
`
`AUTHORS' DISCLOSURES 0F POTENTIAL CONFLICTS
`OF INTEREST
`
`1.0
`0.9
`0.8
`0.7
`0.6
`0.5
`0.4
`0.3
`0.2
`0.1
`
`o
`
`B
`
`a\°
`
`E.
`
`2E:
`
`3
`U)
`Ex.
`u)
`
`>D
`
` I I 'r— I
`
`
`I
`60
`80
`100
`120
`140
`iéo
`180
`
`260
`
`Time(weeks)
`11
`5
`
` I—I —T— I
`
`I
`r
`,—
`180
`200
`40
`80
`100
`120
`140
`160
`
`Time(weel<s)
`
`No. at risk 41
`
`O
`
`
`
`PatientsWithCompleteorPartial
`
`Response(%)
`
`1.0
`0.9
`0.8
`0.7
`0.6
`0.5
`0.4
`0.3
`0.2
`0.1
`
`o
`
`20
`
`No. at risk
`
`(N .= 41).
`(B) Overall survival
`(A) Time to progression (N = 41).
`ig 2.
`Duration of response (in all patients with complete and partial responses).
`
`(C)
`
`and cyclophosphamide together with bevacizumab every 14 days.
`Baseline numbers of CECs correlated with response and outcome,
`which suggests a role for baseline CECs in predicting response to
`such therapy.38
`Another single—arm, phase II trial explored the role of daily oral
`cyclophosphamide and biweekly bevacizumab in patients with meta—
`static ovarian cancer who had received prior treatment with a
`platinum-containing regimen. A PR rate of 24% in 70 patients sug-
`gested activity for this regimen. Response did not correspond to serial
`levels of VEGF, VEGFR, or thrombospondin 1.39
`
`Although all authors completed the disclosure declaration, the following
`author(s) indicated a financial or other interest that is relevant to the subject
`matter under consideration in this article. Certain relationships marked
`with a “U” are those for which no compensation was received; those
`relationships marked with a “C” were compensated. For a detailed
`description of the disclosure categories, orfor more information about
`ASCO’s conflict ofinterest policy, please refer to the Author Disclosure
`Declaration and the Disclosures ofPotential Conflicts ofInterest section in
`Information for Contributors.
`Employment or Leadership Position: None Consultant or Advisory
`Role: Kathleen I. Pritchard, Aventis (C), Pfizer (C), Roche (C),
`
`728
`
`© 2009 by American Society of Clinical Oncology
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubsorg by 64.125.175.43 on December 7, 2017 from 064.125.175.043
`Copyright © 2017 American Society of Clinical OncoloEV. All rights reserved.
`
`
`
`Metronomic DalCMP for Metastatic Breast Cancer
`
`Pharmacia (C), Ortho Biotech (C), YM Biosciences (C), Biomira (C)
`Stock Ownership: None Honoraria: Kathleen I. Pritchard, Aventis,
`AstraZeneca, Pharmacia, Pfizer Research Funding: None Expert
`Testimony: Kathleen I. Pritchard, Aventis (C), AstraZeneca (C) Other
`Remuneration: None
`
`AUTHOR CONTRIBUTIONS
`
`Conception and design: Nan Soon Wong, Gerrit E. DeBoer, Urban
`Emmenegger, Kathleen I. Pritchard
`Financial support: Kathleen I. Pritchard
`Administrative support: Kathleen I. Pritchard
`Provision of study materials or patients: Nan Soon Wong, Susan Dent,
`Maureen E. Trudeau, Gerrit E. DeBoer, Urban Emmenegger, Joyce
`Slingerland, Sandra Gardner, Kathleen I. Pritchard
`
`Collection and assembly of data: Nan Soon Wong, Mark Clemons,
`Shailendra Verma, Susan Dent, Maureen E. Trudeau, Kathie Roche,
`Gerrit E. DeBoer, Donald J.A. Sutherland, Urban Emmenegger, Joyce
`Slingerland, Sandra Gardner, Kathleen I. Pritchard
`Data analysis and interpretation: Shailendra Verma, Susan Dent,
`Maureen E. Trudeau, Kathie Roche, John Ebos, Robert Kerbel, Gerrit E.
`DeBoer, Donald J.A. Sutherland, Urban Emmenegger, Joyce Slingerland,
`Sandra Gardner, Kathleen I. P