UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

CD20 monoclonal antibody therapy for B-cell lymphona

van der Kolk, L.E.

Link to publication

Citation for published version (APA):van der Kolk, L. E. (2001). CD20 monoclonal antibody therapy for B-cell lymphona.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 21 Oct 2020

CC h a pi t e r r

Treatmentt of relapsed B-cell non-Hodgkin's lymphomaa with a combination of chimeric

anti-CD200 monoclonal antibodies (rituximab) and G-CSF:: final report on safety and efficacy

L.E.. van der Kolk1, A.J. Grillo-López2, J.W. Baars3 and M.H.J, van Oers1

departmentt of Hematology, Academic Medical Center, Amsterdam, The Netherlands; 2IDECC Pharmaceuticals Corp., San Diego, CA, USA and department of Medical Oncology,

Antonii van Leeuwenhoek Hospital/Netherlands Cancer Institute Amsterdam, The Netherlands

Submittedd for publication

Possiblee mechanisms of action of the chimeric CD20 monoclonal antibody IDEC-C2B8

(rituximab)) involve complement- and antibody-dependent cellular cytotoxicity (ADCC).

Becausee granulocyte colony-stimulating factor (G-CSF) greatly enhances the cytotoxicity

off neutrophils in ADCC, the clinical efficacy of rituximab might be enhanced by the

additionn of G-CSF. In a phase l/lI clinical trial we investigated the safety and efficacy of

thee combination of rituximab and G-CSF (5u.g/kg/day, administered for three days,

startingg two days before each infusion) in 26 relapsed low-grade non-Hodgkin's

lymphomaa (NHL) patients. Adverse events occurred almost exclusively during the first

infusionn and mainly consisted of (grade l/ll) fever, chills and allergic reactions. Toxicity

wass comparable to that reported for rituximab monotherapy. In the phase II (375

mg/m22 rituximab + G-CSF), 19 patients were evaluable for efficacy. The response rate

wass 42% (8/19; 95% CI 20%-67%). The percentage of complete remissions (CR) and

partiall remissions (PR) were 26% (5/19) and 16% (3/19) respectively. The median

durationn of response has not been reached at a median follow-up of 23 months. Only

onee patient who achieved a CR showed progressive disease (after 18 months), the

otherr 4 patients are still in CR (for 13+, 24+, 25+, 25+ months). We conclude that the

combinationn of rituximab and G-CSF is well tolerated. Although the overall response

ratee seems comparable to that reported for rituximab monotherapy, CR rate is high

andd remission duration in this pilot phase II study is remarkably long. Randomized

comparisonn with rituximab monotherapy should substantiate this promising finding.

Safetyy and efficacy of the combination of rituximab and G-CSF

Introductio n n

Despitee the development of new (chemo-)therapy regimens, the median survival time in

low-gradee non-Hodgkin's lymphoma (NHL) has not changed over the past decades and

curee remains elusive. Thus, the need for new treatment modalities has not abated.

Chimericc anti-CD20 mAbs (IDEC-C2B8, rituximab) have been shown to be a promising new

option.133 It has been demonstrated in B cell-lines in vitro that rituximab can induce

complement-dependentt cytotoxicity (CDC)*6 and antibody-dependent cellular cytotoxicity

(ADCC),, the latter with mononuclear cells as effector cells.4 Furthermore, ligation of CD20

inhibitss B-cell proliferation7 and it has been described that CD20 mAbs can induce apoptosis

inn B-cell-lines, which is enhanced after crosslinking. ^

Recently,, the importance of FcyR-mediated mechanisms for in vivo cytotoxicity of

monoclonall antibodies, including rituximab, was clearly demonstrated.11 Furthermore, the

absolutee number of NK cells prior to onset of treatment with rituximab was found to be

significantlyy correlated to the probability of response to rituximab.12 Hence, although the

exactt mechanism of action of the CD20 mAb in vivo has not been established yet, these

studiess support an important role for Fcy-receptor dependent mechanisms in the efficacy

off rituximab treatment.

Inn a clinical trial in 166 relapsed low-grade NHL-patients treated with 375 mg/m2 rituximab

weeklyy x4, the overall response rate was 48% and the median time to progression 13

months.33 Thus, although clinical results with rituximab as single agent therapy are

encouraging,, they might still be improved. Therefore, several strategies to improve the

clinicall activity of rituximab are currently being investigated.1317

AA possibility to improve the efficacy of unconjugated mAbs might be to enhance the efficacy

off the effector cells involved in ADCC, one of the possible anti-tumor mechanisms of

monoclonall antibodies. Granulocyte colony-stimulating factor (G-CSF) greatly enhances

thee cytotoxic capacity of neutrophils in ADCC-assays.1820 It has been demonstrated in vitro

ass well as in vivo in healthy volunteers, that G-CSF induces the expression of Fcf receptor

typee I (FcyRI) on neutrophils via an effect on myeloid precursor cells.19-21 This FcyR appeared

too be the main FcyR involved in neutrophil-mediated ADCC-assays.18;19 Furthermore, G-CSF

administrationn leads to a large increase in the number of circulating (FcyRI-positive)

neutrophils.. Therefore, adding G-CSF to rituximab therapy could theoretically enhance the

efficacyy of rituximab by exploiting the mechanism of ADCC.

Wee performed a phase l/ll clinical trial to evaluate the safety (phase I) and efficacy (phase

II)) of the combination of rituximab and G-CSF. We hereby present the final data on safety

andd efficacy of this combination.

23 3

Chapterr 2

Material ss and Method s

Studyy design Thiss was an open label, single arm phase l/ll study. Patients were enrolled and treated in

thee Academic Medical Center and the Antoni van Leeuwenhoek Hospital, Amsterdam, the

Netherlands.. The study was approved by the ethics committee at each center and performed

accordingg to the guidelines of the Declaration of Helsinki.

Eachh patient received a total of 4 weekly intravenous (iv) doses of rituximab (IDEC-C2B8,

Rituxan®,, Mabthera®, IDEC Pharmaceuticals, San Diego, CA, USA) in combination with a

standardd dose of G-CSF (filgrastim, Neupogen; Amgen, Thousand Oakes, Ca, USA; 5u.g/

kg/dayy subcutaneously) administered for three days, starting two days before each infusion

off rituximab (fig. 1).

Fig.. 1 Treatmen t schedul e Patientss were treated with the combination of rituximabb and G-CSF. Rituximab was given weekly forr 4 weeks. G-CSF (5 ug/kg/day) was administeredd for 3 days, starting 2 days before eachh infusion of rituximab (i.e., the third injection off G-CSF was administered several hours before rituximabb infusion). Follow-up was at 1,2 and 3 months. .

Thee phase I part of the study consisted of a dose escalation of rituximab, in combination

withh the standard dose of G-CSF. Cohorts of three patients were entered at each dose level

off rituximab (125, 250 or 375 mg/m2 rituximab weekly x4). Dose escalation was allowed

onlyy in the absence of serious toxicity, defined as grade 3 renal toxicity or any grade 4

drug-relatedd toxicity according to the National Cancer Institute's Adult's Toxicity Criteria.

Rituximabb was administered according to the guidelines described in the investigational

drugg brochure.22 In brief, the first infusion was started at 50 mg/hr for the first hour and, if

welll tolerated, the infusion rate was escalated with increments of 50 mg/hr every 30 min,

too a maximum of 300 mg/hr. Subsequent infusions were started at 100 mg/hr for the first

hour,, and increased with 100 mg/hr increments at 30 min-intervals, to a maximum of 400

mg/hr.. Infusion was discontinued in case of hypotension (drop in systolic blood pressure

off >30 mm Hg), rigors, bronchospasm or the occurrence of mucosal congestion or edema.

Iff necessary, medication (consisting of an antihistamine (clemastine 2 mg iv), antipyretic

(paracetamoll 1000 mg) and/or corticosteroids (25 mg prednison iv) was administered.

Afterr resolution of the side effects, infusions were resumed at half of the previous rate, and,

whenn tolerated, increased according to the described schedule.

weekk IV

rituximab b

G-CSF F ttftt ff ttftf l

24 4

Safetyy and efficacy of the combination of rituximab and G-CSF

Afterr evaluation of the phase I, the dose of rituximab for the phase II study was established

att 375 mg/m2. In order to prevent the 'allergic reactions' and fever, which were observed

frequentlyy during the phase I (see 'Results'), it was decided to give premedication during

thee phase II, consisting of clemastine (2 mg iv) and paracetamol (1000 mg). Corticosteroids

weree never given as premedication. Since in phase I side effects were almost exclusively

observedd during the first infusion, premedication was only given before the first infusion.

Patients s Adultss (>18 years) were eligible if they had measurable relapse or progression of histologically

confirmedd CD20-positive B-cell lymphoma. In the phase I part, patients with low and

intermediatee grade NHL were included (Working Formulation (WF) A-D). In the phase II part,

onlyy low-grade NHL patients were included (WF A-C).

Alll patients had to meet the following inclusion criteria: expected survival of > 3 months;

prestudyy performance status of 0-2 according to the World Health Organization (WHO) scale;

relapsee or progression after at least one and no more than three prior systemic therapies;

sero-negativee for human immunodeficiency virus (HIV) and hepatitis-B surface antigen (HBsAg);

serumm IgG > 600 mg/dl; hemoglobin (Hb) > 5 mmol/L; white blood cell count (WBC) > 3.0

x109/L;; absolute granulocyte count > 1.5 x109/L; platelet count (Pit) > 75 x109/L; circulating

tumorr cells < 0.5 xl 09/L. The number of circulating tumor cells allowed was initially very low

(cellss < 0.5 x109/L) because of a possible relation between circulating tumor cells and side

effects.. During the study it became clear that side effects were generally mild and therefore

thiss inclusion criterium was changed to 'circulating tumor cells < 2.5 xl 09/L'.

Patientss were ineligible if they had received other lymphoma treatment within three weeks

priorr to the scheduled first treatment; if they had a bilirubin level of > 1.5 mg/dL (26 jimol/L);

alkalinee phosphatase level of > 2x upper range of normal; AST (SGOT) level of > 2x upper

rangee of normal; serum creatinine level of > 2.0 mg/dL (177 u.mol/L) or if they had a serious

non-malignantt disease. All patients gave written informed consent before treatment was

initiated. .

Clinicall parameters Monitoringg of safety was performed throughout the study and included medical history,

vitall signs, analysis of renal, hepatic and hematologic parameters and analysis of Ig levels.

Toxicityy was evaluated according to the National Cancer Institute's Adult's Toxicity Criteria.

Responsee criteria were as follows: a complete remission (CR) required complete

disappearancee of all clinically detectable disease, including bone marrow infiltration, for at

leastt 28 days; partial remission (PR) was defined as a > 50% decrease in overall tumor size

25 5

Chapterr 2

withoutt any evidence of disease progression for at least 28 days; stable disease (SD) was

definedd as <50% decrease or <25% increase in overall tumor size, without any evidence

off progression; progressive disease (PD) was defined as >25% increase in overall tumor

size,, or >50% increase in any single lesion, or the appearance of new lesions.

Durationn of response was measured from initial observation of response until the last date

thatt the measurements satisfied the criteria of the response. Time to progression was

measuredd from first study treatment until the first date when progressive disease was

documented. .

Immunophenotypicall analysis Mabss against CD11b (CLB-mon-gran/1, B2), CD13 (CLB-mon-gran/2), CD16 (CLB-

FcRgran/1,, 5D2), CD45 (CLB-T200/1, 15D9), CD64 (10.1), CD66b (CLB-B13.9) and the

irrelevantt control antibodies lgG1 and lgG2a were from the CLB (Amsterdam, The

Netherlands).. FITC-conjugated rabbit antknouse-lg (F(ab')2 fragments) and Kappa and

Lambdaa mAbs were from Dako (Glostrup, Denmark). Mean fluorescence intensity (MFI)

wass measured by flow cytometry (FACS-scan, Becton Dickinson, San Jose, CA, USA) and

dataa were always corrected for the MFI in the presence of irrelevant control antibody.

Elastase/lactoferrin n Plasmaa levels of elastase-a1-antitrypsin complexes (further referred to as elastase) and

lactoferrinn were measured with radioimmunoassays (RIA) as described before.23 The levels

aree expressed as ng/ml using preformed complexes and purified lactoferrin as standards

respectively.. Levels of elastase and lactoferrin in healthy individuals are below 100 ng/ml

andd 400 ng/ml respectively.23

Statistics s Statisticall analysis was performed using the SPSS statistical package (SPSS for Windows, version

8.0).. Data are expressed as means SEM, unless indicated otherwise. Serum concentrations

weree compared using the Wilcoxon Signed Ranks test. Correlations were assessed using

Pearson'ss correlation coefficient. Differences in response rates between groups were tested

forr statistical significance using Fisher's exact test (two-sided). The duration of response was

assessedd using the method of Kaplan and Meier. A P-value < 0.05 was considered significant.

26 6

Safetyy and efficacy of the combination of rituximab and G-CSF

Result s s

Patientt characteristics Ninee patients were treated in the phase I part of the study, three patients at each consecutive

dosee level. Patient characteristics are listed in table 1. The median age was 53 years (range

32-599 yrs) and the histologic subtype was low grade NHL in 7 patients and intermediate

gradee NHL in 2 patients. In the phase II, 17 patients have been treated. The median age

waswas 53 years (range 27-75 yrs) and all patients had a low-grade NHL (table 1). The median

numberr of prior systemic therapies was 2, and regimens included CVP (n=19 patients),

CHOPP (n=8), chlorambucil (n=5) and fludarabin (n=6). One patient had undergone high

dosee chemotherapy with stem cell rescue 5 months before study entry.

Tablee 1. Patien t characteristic s

Agee at stud y entr y (year) median n range e

Classificatio nn (WF) A A B B C C D D E E

No.. of prio r systemi c regimens** * 1 1 2 2 3 3

Stagee at diagnosi s IE E IIA A IIB B IMA A 1MB B IVA A

IVB B

IVE E

phasee 1 n=9 9

53 3 32-59 9

1* * 5 5 0 0 2 2 1 1

3 3 2 2 3 3

1 1 1 1 0 0 1 1 0 0 3 3

2 2

1 1

phasee II n=17 7

53 3 27-75 5

4** * 11 1 2 2 0 0 0 0

10 0 4 4 3 3

0 0 2 2 1 1 2 2 0 0 11 1

1 1

0 0

** maltoma; ** 3 x immunocytoma, 1 x marginal zone lymphoma, according to the Real/Kiel classification.. * * * One patient (phase I) had received 4 prior regimens

27 7

Chapterr 2

Toxicity y Alll but one patient received the 12 injections of G-CSF according to the study protocol.

Administrationn of G-CSF did not induce significant toxicity. However, there were patients

thatt experienced mild bone pain or flu-like symptoms. In those patients it was advised to

takee paracetamol (1000 mg) before G-C5F administration. One patient had experienced

gradee II side effects (bone pain) during the first cycle of G-CSF. At her request, she did not

receivee G-CSF preceding the subsequent rituximab infusions.

Thee nine patients included in the phase I all completed the four cycles of rituximab treatment

andd were evaluable for safety. No premedication was given. Adverse events are summarized

inn table 2. Toxicity during rituximab treatment was experienced by 8 (out of 9) patients and

consistedd mainly of infusion-related adverse events (total number of adverse events =30).

Thee most frequent adverse events were fever and 'allergic reactions' consisting of rhinitis,

sneezing,, itching of the oropharynx and a sensation of swelling of the throat. There was no

apparentt relation between the dose of rituximab and the amount or severity of the adverse

eventss (see table 1). The infusion-related adverse events typically occurred one to two

hourss after onset of the infusion. Most events were classified as mild to moderate (toxicity

gradess 1-2) and occurred during the first infusion; only few patients experienced adverse

eventss during subsequent infusions. One patient developed a viral stomatitis (2 weeks after

completionn of rituximab therapy), which was complicated by a bacterial superinfection.

Tablee 2. Infusio n relate d advers e events

rituximabb dose (mg/m

fever r 'allergicc reaction' headache e pruritis s chills s pain** * rash h nausea a urticaria a hypotension n other r totall adverse events

2x4) )

'Allergicc reaction' refers to a

125 5 (n-3) )

3 3 4 4 2 2 2 2 1 1 1 1 1 1 1 1 0 0 0 0 0 0 15 5

phasee I

250 0 (n-3) )

2 2 2 2 0 0 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 8 8

375 5 (n-3) )

2 2 2 2 1 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 1 7 7

numberr of reactions including rhinitis,

totall phase I group* *

7(4) ) 8(4) ) 3(2) ) 3(1) ) 2(2) )

KD D KD D 1(1) ) 1(1) ) 1(1) ) 2(0) )

30(18) )

sneezing,, itching of

phasee II

375 5 (n-17) )

19(13) ) 9(6) ) 4(2) ) 1(0) ) 7(4) ) 1(1) ) 2(0) ) 1(1) )

----

17(9) ) 611 (36)

oropharynxx and aa sensation of swelling of the throat. * The number of adverse events during the first infusion is given inn parenthesis. ** Pain at tumor sites

28 8

Safetyy and efficacy of the combination of rituximab and G-CSF

Thiss infection resolved upon treatment with acyclovir and antibiotics.

Thoughh no serious infusion-related toxicity was observed during the phase I, it was decided

too treat patients participating in the phase 11 study prophylactically with clemastine (2 mg iv)

andd paracetamol (1000 mg) in order to attenuate the allergic reactions that were frequently

observedd during the phase I study. Infusion related side effects observed in the phase II

were'comparablee to those documented in the phase I. However, the allergic reactions as

describedd in the phase I patients appeared to occur less frequently in these patients (9/61

duringg phase II versus 8/30 during phase I) and were clinically less pronounced than during

phasee I. Only mild hematologic toxicity was observed (grade 1-2) and parameters usually

recoveredd at follow up. No renal or hepatic toxicity was observed (data not shown).

Neutrophill activation upon G-CSF-administration Neutrophill activation was assessed 1) by analysis of phenotypic changes of neutrophils and

2)) by measuring plasma levels of elastase and lactoferrin as markers for neutrophil

degranulation.. The rationale of adding G-CSF to the treatment with rituximab was to induce

thee expression of the FcyRI on the circulating neutrophils, while at the same time increasing

theirr number. As shown in fig. 2, directly prior to the start of the first rituximab infusion the

numberr of neutrophils was strongly increased (from 4 0.4 x109/L to 19 + 2 x109/U

PO.001).. Moreover, FcyRI-expression on neutrophils rose from 26 9 (MFI) at baseline to

688 + 8 (P=0.001) after three days of G-CSF. Three days after treatment, the mean expression

levell of FcyRI on circulating neutrophils was still elevated. The increase in neutrophil count

]] neutrophils -CD64 4

xx 40 C C

88 3 0

ff 20 H

ii 10-

O O

A A £L L

X X ** **

J2 2 -488 -1 72 -hrs-

weekk I -488 -1 72

weekk IV

100 0

800 £

600 ^r 10 0

400 8

20 0

0 0

400-, ,

300--

200 0

-CDIIb-- -CD13-- -CD16 6

OO 100 ^^ \7.

2500 0

2000 0

-- 1500 2 to o

10000 5 u u

500 0 0 0

-488 -1 72 -hrs- -48 -1 72 weekk I week IV

Fig.. 2 Neutrophi l count s and expressio n of activatio n marker s A)) Absolute number of neutrophils (x109/L) and expression of CD64 (FcyRI) and B) expression of CD111 b, CD13 and CD16 (FcyRIII) during treatment with rituximab and G-CSF. Values are mean SEM off 19 patients (i.e. all patients that received 375 mg/m2 rituximab in combination with G-CSF). Timee points are related to rituximab infusion: -48: 48 hrs before rituximab infusion, before first administrationn of G-CSF; - 1 : 1 hour before rituximab infusion, after 3 injections of G-CSF; 72: 72 hourss after both rituximab infusion and the third G-CSF injection. Asterisks indicate a significant differencee (*: PO.05, **: PO.01) as compared to baseline value (-48 hours).

29 9

Chapterr 2

andd induction of FcyRI-expression by G-CSF showed a comparable pattern during the first

weekk and the fourth week of treatment (fig. 2A).

CD11bb (C3bi receptor), CD13 (aminopeptidase N), CD45 (membrane associated tyrosine

kinase)) and CD66b (CEA/like antigen, mainly localized in the specific granules) are antigens

consideredd to be associated with neutrophil activation. Expression levels of these antigens

weree determined by flow cytometry, at different timepoints during in the first and fourth

weekk of treatment and all values were compared to baseline levels (week I, timepoint -48 hrs,

seee legend fig. 2). CD13, CD16 (FcyRIII) and CD11b decreased upon G-CSF administration

(fig.. 2B). However, expression returned to normal at day three after rituximab treatment.

Theree was no significant change in the mean expression of CD45 and CD67 (not shown).

160-, ,

120 0

»» 80

40 0

0 0 o—<̂

-488 -1 72 -hrs- -48 -1 72

weekk I week III

800 0 1 1 ~3ii 600 £ £

oo 400 O O Ü Ü .22 200 4 ^ ^ ^

488 -1 72 -hrs- -48 -1 72

weekk I week III

4000 i

300--

200--

100--

n n

—0—— elastase—H-

i i

^ ^

H H

-- lactoferrin

I I

P-*^--r* r*

1000 0

8000 "E

00 30 60 90 180 300

minn after onset first infusion

600 0

4000 €

200 0

Fig.. 3 Levels of elastas e and lactoferri n A,, B) Elastase (A) and lactoferrin (B) measuredd at different time points (see legendd fig. 2). Values are mean SEM of 7 patients.. C) Elastase and lactoferrin measuredd in serial samples during the first infusionn of rituximab. Timepoints are minutess after start infusion. Values are meann SEM of 5 patients. Asteriskss indicate a significant difference (*:: PO.05, **: PO.01) as compared to baselinee value (A, B:-48; C: t= 0).

Thee concentration of elastase was 49 + 7 ng/ml at baseline (t=- 48hrs). After 3 injections of

G-CSF,, the concentration increased to 129 20 ng/ml (t=-1 hr, just before rituximab infusion;

P=0.018).. Three days later (t= +72 hours), levels had returned to normal (<100 ng/ml). The

samee pattern was observed during the fourth week of treatment (fig. 3A).

Lactoferrinn levels rose from 179 35 ng/ml at baseline, to 395 2 ng/ml after 3 days of

G-CSFF (t=-1 hr; not significant (NS)), which is just within normal limits. At 72 hours after G-

30 0

Safetyy and efficacy of the combination of rituximab and G-CSF

CSF,, lactoferrin levels further increased to 432 100 ng/ml (t=72 hrs, NS) (fig. 3B). During

thee fourth week of treatment, lactoferrin levels were significantly increased when compared

too baseline values (p<0.05 at all time points, fig. 3B). In two control patients, treated with

rituximabb (375 mg/m2) only, levels of elastase or lactoferrin did not change at these time

points,, indicating that the observed increases were due to G-CSF pretreatment.

Neutrophill activation upon rituximab infusion Too investigate whether neutrophils were additionally activated during infusion of rituximab

wee assessed kinetics of plasma levels of elastase and lactoferrin during the first infusion of the

antibodyy in five patients. At the start of the infusion, levels were already slightly increased due

too G-CSF pretreatment (see previous paragraph). Upon infusion of rituximab, elastase rose

fromm 165 38 ng/ml to a maximum of 301 36 ng/ml at 180 minutes; lactoferrin rose from

2544 36 ng/ml at baseline to a maximum of 630 172 at 90 minutes. At the end of the

infusion,, elastase and lactoferrin had almost returned to pre-infusion values (fig. 3C).

Response e Twentyy patients have been treated with the highest dose of rituximab (375 mg/m2) in

combinationn with G-CSF, 3 in the phase I and 17 in the phase II part of the study. Since one

patientt did not receive all G-CSF injections, 19 patients were evaluable for response (table 3).

Thee response rate was 42% (95%CI 20%-67%) (40% on intent-to treat basis). The

percentagee of complete remissions (CR) was 26% (5/19) and partial remissions (PR) was

16%% (3/19). In two patients, the complete remission could not be confirmed by bone

marroww biopsy (due to patient refusal). The median duration of response has not been

reachedd at a median follow-up of 23 months (table 4). Only one patient who achieved a CR

relapsedd (after 18 months remission duration), the other 4 patients are still in CR. The duration

off partial remissions was shorter: 4, 10 and 4+-months respectively. In 4 (from 8) patients,

Tablee 3. Respons e to the combinatio n of rituxima b (3755 mg/m 2x4) and G-CSF

response e

CR** * PR R SD D PD D

ORR (CR+PR)

n=19* *

5 5 3 3 7 7 4 4

42%% (8/19)

** of 20 patients treated, 19 patients were evaluable for efficacy. ** * 2 CR's are not confirmed by bone marrow biopsy

31 1

Chapterr 2

thee duration of response to rituximab treatment was already considerably longer than the

responsee to last systemic regimen (table 4).

Althoughh the sample size is rather small, factors that have previously been reported to

influencee the response to rituximab treatment were analyzed in our patient group as well.

Patientss that achieved a CR or PR to their last prior chemotherapy regimen (n=11) seemed

too have a better response to rituximab therapy than patients that did not respond to last

priorr therapy (n-8) (response rate 63% (7/11) vs 13% (1/8) respectively, P=0.059). The

responsee rate in patients who had previously received only one systemic treatment (n=10)

wass similar to the response in those who had received >1 regimens (n=9) (response rate

50%% (5/10) vs 33% (3/9) respectively, P=0.65). No difference in response rates were

observedd between patients with bulky disease (defined as lesions >7 cm) versus (vs) patients

withoutt bulky disease (P=0.17), patients with bone marrow infiltration vs without bone

marroww infiltration (P=0.37), and patients with extranodal lesions vs without extranodal

lesionss (P=1.0). In previous studies, WF A lymphoma was found to be correlated with a

lowerr response to rituximab treatment, as compared to WF B and/or WF C lymphoma.324

Amongg the 19 evaluable patients in the present study, there were 4 with WF A lymphoma

(33 patients with immunocytoma (Kiel classification) and one with marginal zone lymphoma

(Reall classification)). These patients were all non responders.

Tablee 4. Duratio n of respons e to rituxima b treatmen t

patientt response and duration of response* to rituximab*** last prior regimen***

11 1

12 2

14 4

15 5

16 6

17 7

23 3

28 8

PR R

CR R

CR R

CR R

CR R

PR R

CR R

PR R

4(6) )

25++ (29+)

18(24) )

25++ (27+)

24++ (26+)

9(12) )

13+(16+) )

4++ (6+)

PR R PR R

CR R

CR R

CR R

NC C

CR R

CR R

27 7 2 2

13 3

3 3

12 2

NA A

47 7

26 6

Abbreviations:: CR: complete remission; PR: partial remission; NC: no change; NA:: not applicable. * in months ** the time to progression is given in parenthesiss * * * last systemic chemotherapy regimen

32 2

Safetyy and efficacy of the combination of rituximab and G-CSF

Discussio n n

Inn the present study, we evaluated safety and efficacy of the combination of rituximab and

G-CSFF in relapsed low-grade NHL patients.

Sidee effects of G-CSF were mild and consisted mainly of bone pain and flu-like symptoms,

whichh is in line with previous studies on G-CSF administration.25 Side effects during rituximab-

infusionn were mild (toxicity grade 1-2) and included mainly fever and 'allergic reactions',

consistingg of sneezing, rhinitis and itching of the oropharynx. Thus, the combination therapy

off rituximab and G-CSF was well tolerated and toxicity appeared to be similar to that reported

forr rituximab monotherapy.13

Thee response rate was 42% (95% CI 20%-67%), and comparable to the response rate

achievedd with rituximab monotherapy.3;24 Interestingly, the CR rate was high and the duration

off responses was remarkably long. At a median follow up of 23 months, 5 (from 8)

responderss were still in remission, 3 of them for >24 months. Thus, the median time to

progressionn has not been reached yet. Duration of CR appeared to be longer than of PR

and,, furthermore, in 4 (from 5) patients that achieved a CR the duration was already

considerablyy longer than remission duration achieved on last previous systemic therapy.

Thus,, although the overall response rate was not increased, addition of G-CSF to rituximab

mayy be beneficial by increasing response duration.

Still,, results were not as favorable as hoped for. Theoretically, possible explanations might

be:: 1) neutrophils did not express FcyRI; 2) FcyRI-positive neutrophils were not capable

off mediating CD20-dependent ADCC; 3) competition of rituximab with circulating

monomericc IgG for the ligand binding site of FcyRI; 4) FcyRI-positive neutrophils did not

reachh tumor sites.

Thee expression of FcyRI on neutrophils was significantly increased after three injections of

G-CSFF (i.e. just prior to rituximab infusion) (fig. 2). Previously, the capacity of CD20 mAbs in

mediatingg ADCC has been questioned.20 We found however, that chimeric CD20 mAbs

weree capable of inducing ADCC in B cells with G-CSF-primed neutrophils as effector cells,

whereass unstimulated neutrophils could not induce CD20-dependent ADCC (van der Kolk

ett al, manuscript in preparation). Thus, as intended, G-CSF administration resulted in FcyRI-

positive,, cytotoxic neutrophils.

Itt has been suggested that circulating monomeric IgG inhibits the function of FcyRI in vivo.26

However,, several observations challenge this suggestion. First, from a study by Heijnen et

al277 it can be deduced, that although a competition between chimeric antibodies and

endogenouss IgG for FcyRI might exist, this can be overcome when adequate amounts of

antibodyy are administered. Notably, concentrations of rituximab reached during treatment

33 3

Chapterr 2

aree high, with maximum levels up to 660 u.g/ml_, easily exceeding the maximum

concentrationn used in the described study.2;27 More important however is the observation

thatt FcyR Ill-deficient mice still develop autoimmune hemolytic anemia (AIHA) upon

administrationn of monoclonal antibodies against red blood cell antigens in vivo. This

AIHAA was found to be mediated by FcyRI (and not FcyRII), indicating the functional activity

off FcyRI in vivo.2S:29 Taking these arguments together, we do not think that competition

withh circulating IgG for FcyRl-binding has hampered the clinical efficacy of rituximab in

thee present study.

Inn order to emigrate to lymphatic tissue or sites of inflammation, neutrophils need to be

attractedd (by chemoattractants, e.g. cytokines, histamine, C5a) and migratory capacity should

bee adequate. Though B-cells in lymph nodes were coated with rituximab (data not shown)

ass was also observed in previous studies4130, the degree of inflammation induced by rituximab

att tumor sites may be insufficient to initiate neutrophil migration. Furthermore, compared

too unstimulated neutrophils, in vivo G-CSF-stimulated neutrophils show certain differences

thatt may influence the migratory capacity in vivo. First, expression of L-selectin (CD62L), an

adhesionn molecule playing an important role in the initial interaction between neutrophils

andd endothelial cells (rolling)3133, and found to be essential for leukocyte recruitment to

lymphaticc tissue and inflammatory sites34"37, was shown to be downregulated after in vivo

G-CSF-administration.3M11 Second, upon G-CSF administration we found a decrease in the

expressionn of CD 11 b, a (32-integrin that plays an important role in the adhesion of neutrophils

too endothelium (fig. 2).31;32 This was in contrast to previous reports in which upon in vivo G-

CSF-administrationn an increase in CD11b was found42-44, accompanied by an increase in

adhesionn of neutrophils to various cells and coatings.42143 Although it is known that

conformationall changes in integrins may be more important for adhesion than the actual

expressionn level45;46, and G-CSF-stimulated neutrophils were found to have a normal capacity

too upregulate CD11b in response to activating stimuli47, the decrease in CD11b may still

havee influenced the migration of neutrophils in this study. Finally, the directed migration

(chemotaxis)) of in vivo G-CSF-stimulated neutrophils was found to be decreased.42;47:48

Inn the present study, lymph node biopsies were taken one day after the last rituximab

infusionn in three patients. Although the lymphocytes were coated with rituximab, only few

neutrophilss were observed (data not shown). Thus, at least at the described time point,

neutrophilss did not migrate to tumor sites in large numbers. This might suggest that

insufficientt migration has hampered the beneficial effects of G-CSF. However, since only

threee biopsies were taken, we can not draw definite conclusions as to this possibility.

Respiratoryy side effects during treatment with monoclonal antibodies have been attributed

too the accumulation of activated neutrophils in the lung vasculature.49 Interestingly, although

ourr patients had very high numbers of neutrophils that subsequently became activated

34 4

Safetyy and efficacy of the combination of rituximab and G-CSF

duringg rituximab infusion (fig. 3C), only one patient showed respiratory problems. In analogy

too the study by Pajkrt et al50, G-CSF-induced downregulation of L-selectin may have prevented

neutrophilss from accumulation in the lungs. Thus, the same mechanism thatt theoretically

mightt have decreased the beneficial effect of G-CSF may at the same time have protected

patientss for potential side effects.

Recently,, results of a clinical trial evaluating the safety and efficacy of the combination of

rituximabb and interferon-a-2a (IFN-a) were published.51 IFN-a is known to stimulate the

abilityy of NK<ells in mediating ADCC. Thus, combining rituximab with G-CSF or IFN-a is a

similarr approach intended to increase the efficacy of rituximab.

Thee overall response rate of the combination of rituximab and IFN-a was 45% (i.e. similar

too rituximab monotherapy), but response duration appeared to be longer than for rituximab

monotherapy.. However, side effects observed during treatment with rituximab and IFN-a

weree more severe, including grade 3 adverse events in 12 patients. The majority of the side

effectss were attributed to IFN-a. Hence, although the potential beneficial effects of IFN-a

andd G-CSF on the response duration of rituximab may be similar, the safety profile of the

combinationn with G-CSF is clearly more favorable.

Inn conclusion, the combination of G-CSF and rituximab was well tolerated in relapsed low

gradee NHL-patients. Although the response rate was comparable to that reported for

rituximabb monotherapy, CR rate was high and remission duration in this smal! sized phase II

studyy was remarkably long. Randomized comparison with rituximab monotherapy should

substantiatee this promising finding.

35 5

Chapterr 2

Referenc ee List 1.. Maloney DG, Grillo-López AJ, Bodkin DJ, et al. IDEC-C2B8: results of a phase I multiple-dose trial in

patientss with relapsed non-Hodgkin's lymphoma. J.Clin.Oncol. 1997;15:3266-3274. 2.. Maloney DG, Grillo-López AJ, White CA, et al. IDEGC2B8 (Rituximab) anti-CD20 monoclonal antibody

therapyy in patients with relapsed low-grade non-Hodgkin's lymphoma. Blood. 1997;90:2188-2195. 3.. McLaughlin P, Grillo-López A, Link BK, et al. Rituximab chimeric anti-CD20 monoclonal antibody therapy

forr relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J.Clin.Oncol.. 1998;16:2825-2833.

4.. Reff ME, Carner K, Chambers KS, et al. Depletion of B cells in vivo by a chimeric mouse human monoclonall antibody to CD20. Blood. 1994;83:435-445.

5.. Golay J, Zaffaroni L, Vaccari T, et al. Biologic response of B lymphoma cells to anti-CD20 monoclonal antibodyy rituximab in vitro: CD55 and CD59 regulate complement-mediated cell lysis. Blood. 2000;95:3900-3908. .

6.. Harjunpaa A, Junnikkala S, Meri S. Rituximab (Anti-CD20) therapy of B-cell lymphomas: direct complementt killing is superior to cellular effector mechanisms. ' Scand.J.Immunol. 2000;51:634-641.

7.. Tedder TF, Boyd AW, Freedman AS, Nadler LM, Schlossman SF. The B cell surface molecule B1 is functionallyy linked with B cell activation and differentiation. J.Immunol. 1985;135:973-979.

8.. Hofmeister JK, Cooney D, Coggeshall KM. Clustered CD20 Induced Apoptosis: Src-Family Kinase, the Proximall Regulator of Tyrosine Phosphorylation, Calcium Influx, and Caspase 3-Dependent Apoptosis. Bloodd Cells Mol.Dis. 2000;26:133-143.

9.. Shan D, Ledbetter JA, Press OW. Apoptosis of malignant human B cells by ligation of CD20 with monoclonall antibodies. Blood. 1998;91:1644-1652.

10.. Shan D, Ledbetter JA, Press OW. Signaling events involved in anti-CD20-induced apoptosis of malignant humann B cells. Cancer Immunol.Immunother. 2000;48:673-683.

11 1. Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivo cytoxicity against tumorr targets. Nat.Med. 2000;6:443-446.

11 2. Janakiraman N, McLaughlin P, White CA, Maloney DG, Shen D, Grillo-López A. Rituximab: correlation betweenn effector cells and clinical activity in NHL [abstract]. Blood. 1998;92:337a

13.. Kaminski MS, ZasadnyKR, Francis IR, et al. lodine-131-anti-B1 radioimmunotherapy for B-cell lymphoma. J.CIin.Oncol.. 1996;14:1974-1981.

14.. Liu SY, Eary JF, Petersdorf SH, et al. Follow-up of relapsed B-cell lymphoma patients treated with iodine-131-- labeled anti-CD20 antibody and autologous stem-cell rescue. J.CIin.Oncol. 1998;16:3270-3278. .

11 5. Kimby E, Geisler G, Hagberg H, et al. Rituximab (Mabthera®) as a single agent and in combination with interferon-alpha-2aa as treatment of untreated and first relapse follicular or other low-grade lymphomas. Preliminaryy results of a randomized phase II study (M39035) [abstract!. Blood. 1999;94:4388a

11 6. Kleinman M, Kaplan E, Rhoades D, Zats S, Wechsler J. Sargramostim/rituximab therapy for B-cell non-Hodgkin'ss lymphoma (NHL) and chronic lymphocytic leukemia (CLL) patients: a toxicity profile [abstract]. Blood.. 1999;94:4389a

11 7. Sacchi S, Federico M, Vitolo U, et al. Phase II study of rituximab after priming with IFN in patients with relapsedd LG/F B-cell lymphoma [abstract]. Blood. 1999;94:399a

18.. Valerius T, Repp R, de Wit T, et al. Involvement of the high-affinity receptor for IgG (Fc gamma Rl; CD64)) in enhanced tumor cell cytotoxicity of neutrophils during granulocyte colony-stimulating factor therapy.. Blood. 1993;82:931-939.

19.. Kerst J, de Haas M, van der Schoot C, et al. Recombinant granulocyte colony-stimulating factor administrationn to healthy volunteers: induction of immunophenotypically and functionally altered neutrophilss via an effect on myeloid progenitor cells. Blood. 1993;82:3265-3272.

20.. Elsasser D, Valerius T, Repp R, et al. HLA class II as potential target antigen on malignant B cells for therapyy with bispecific antibodies in combination with granulocyte colony-stimulating factor. Blood. 1996;87:3803-3812. .

21 .. Repp R, Valerius T, Sendler A, et al. Neutrophils express the high affinity receptor for IgG (Fc gamma Rl,, CD64) after in vivo application of recombinant human granulocyte colony- stimulating factor. Blood.. 1991;78:885-889.

22.. Investigational drug brochure rituximab (RituxanTM, Mab Thera, IDEC-C2B8) edition 4, IDEC Pharmaceuticalss Corporation, San Diego, Ca, USA. 1997.

36 6

Safetyy and efficacy of the combination of rituximab and G-CSF

23.. Nuijens JH, Abbink JJ, Wachtfogel YT, et al. Plasma elastase alpha 1-antitrypsin and lactoferrin in sepsis:: evidence for neutrophils as mediators in fatal sepsis. J.Lab.Clin.Med. 1992;119:159-168.

24.. Foran JM, Gupta RK, Cunningham D, et al. A UK multicentre phase II study of rituximab (chimaeric anti-CD200 monoclonal antibody) in patients with follicular lymphoma, with PCR monitoring of molecular response.. Br.J.Haematol. 2000;109:81-88.

25.. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulatingg factor (1). N.Engl.J.Med. 1992;327:28-35.

26.. Deo YM, Graziano RF, Repp R, van de Winkel JG. Clinical significance of IgG Fc receptors and Fc gammaa R-directed immunotherapies. Immunol.Today. 1997;18:127-135.

27.. Heijnen IA, Rijks U, Schiel A, et al. Generation of HER-2/neu-specific cytotoxic neutrophils in vivo: efficientt arming of neutrophils by combined administration of granulocyte colony-stimulating factor andd Fcgamma receptor I bispecific antibodies. J.lmmunol. 1997;159:5629-5639.

28.. Meyer D, Schiller C, Westermann J, et al. FcgammaRIM (CD16)-deficient mice show IgG isotype-dependent protectionn to experimental autoimmune hemolytic anemia. Blood. 1998;92:3997-4002.

29.. Fossati-Jimack L, loan-Facsinay A, Reininger L, et al. Markedly different pathogenicity of four immunoglobulinn G isotype-switch variants of an antierythrocyte autoantibody is based on their capacity too interact in vivo with the low-affinity Fcgamma receptor III. J.Exp.Med. 2000;191:1293-1302.

30.. Maloney DG, Liles TM, Czerwinski DK, et al. Phase I clinical trial using escalating single-dose infusion of chimericc anti-CD20 monoclonal antibody (IDEC-C2B8) in patients with recurrent B-cell lymphoma. Blood. 1994;84:2457-2466. .

31.. Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules. Blood. 1994;84:2068-2101. 32.. Hogg N. Roll, roll, roll your leucocyte gently down the vein.... Immunol.Today. 1992;13:113-115. 33.. Ley K, Tedder TF. Leukocyte interactions with vascular endothelium. New insights into selectin-mediated

attachmentt and rolling. J.Immunol. 1995;155:525-528. 34.. Tedder TF, Steeber DA, Pizcueta P. L-selectin-deficient mice have impaired leukocyte recruitment into

inflammatoryy sites. J.Exp.Med. 1995;181:2259-2264. 35.. Mihelcic D, Schleiffenbaum B, Tedder TF, Sharar SR, Harlan JM, Winn RK. Inhibition of leukocyte L-

selectinn function with a monoclonal antibody attenuates reperfusion injury to the rabbit ear. Blood. 1994;84:2322-2328. .

36.. Steeber DA, Green NE, Sato S, Tedder TF. Lyphocyte migration in L-selectin-deficient mice. Altered subsett migration and aging of the immune system. J.Immunol. 1996;157:1096-1106.

37.. Arbones ML, Ord DC, Ley K, et al. Lymphocyte homing and leukocyte rolling and migration are impaired inn L-selectin-deficient mice. Immunity. 1994;1:247-260.

38.. Ohsaka A, Saionji K. In vivo administration of granulocyte colony-stimulating factor increases the surfacee expression of sialyl-Lewis{x) on neutrophils in healthy volunteers. Acta Haematol. 1998; 100:187-190. .

39.. Stroncek DF, Jaszcz W, Herr GP, Clay ME, McCullough J. Expression of neutrophil antigens after 10 dayss of granulocyte-colony-stimulating factor. Transfusion. 1998;38:663-668.

40.. Ohsaka A, Saionji K, Sato N, Mori T, Ishimoto K, Inamatsu T. Granulocyte colony-stimulating factor down-regulatess the surface expression of the human leucocyte adhesion molecule-1 on human neutrophils inn vitro and in vivo. Br.J.Haematol. 1993;84:574-580.

4 1 .. Spiekermann K, Roesler J, Emmendoerffer A, Eisner J, Welte K. Functional features of neutrophils inducedd by G-CSF and GM-CSF treatment: differential effects and clinical implications. Leukemia. 1997;11:466-478. .

42.. Hoglund M, Hakansson L, Venge P. Effects of in vivo administration of G-CSF on neutrophil functions inn healthy volunteers. Eur.J.Haematol. 1997;58:195-202.

43.. Hakansson L, Hoglund M, Jonsson UB, Torsteinsdottir I, Xu X, Venge P. Effects of in vivo administration off G-CSF on neutrophil and eosinophil adhesion. Br.J.Haematol. 1997;98:603-611.

44.. de Haas M, Kerst JM, van der Schoot C, et al. Granulocyte colony-stimulating factor administration to healthyy volunteers: analysis of the immediate activating effects on circulating neutrophils. Blood. 1994;84:3885-3894. .

45.. Diamond MS, Springer TA. A subpopulation of Mac-1 (CD11b/CD18) molecules mediates neutrophil adhesionn to ICAM-1 and fibrinogen. J.Cell Biol. 1993;120:545-556.

46.. Elemer GS, Edgington TS. Monoclonal antibody to an activation neoepitope of alpha M beta 2 inhibits multiplee alpha M beta 2 functions. J.Immunol. 1994;152:5836-5844.

37 7

Chapterr 2

47.. Leavey PJ, Sellins KS, Thurman G, et al. In vivo treatment with granulocyte colony-stimulating factor resultss in divergent effects on neutrophil functions measured in vitro. Blood. 1998;92:4366-4374.

48.. Price TH, Chatta GS, Dale DC. Effect of recombinant granulocyte colony-stimulating factor on neutrophil kineticss in normal young and elderly humans. Blood. 1996;88:335-340.

49.. Raasveld MH, Bemelman FJ, Schellekens PT, et al. Complement activation during OKT3 treatment: a possiblee explanation for respiratory side effects. Kidney Int. 1993;43:1140-1149.

50.. Pajkrt D, Manten A, van de Poll T, et al. Modulation of cytokine release and neutrophil function by granulocytee colony-stimulating factor during endotoxemia in humans. Blood. 1997;90:1415-1424.;

51.. Davis TA, Maloney DG, Grillo-López AJ, et al. Combination immunotherapy of relapsed or refractory low-gradee or follicular non-Hodgkin's lymphoma with rituximab and interferon-alpha-2a. Clin.Cancer Res.. 2000;6:2644-2652.

38 8