Blood cytokine response of low-dose molgramostim (rhGM-CSF)-treated patients

(1)

MOLGRAMOSTIM (rhGM-CSF)-TREATED PATIENTS

Thomas Hartung,

*

Sonja von Aulock,' Marcus Freitag," Stefan Hoxtermann,"

Markus Stucker," Klaus Hoffmann,:' PeterAltmeyer.? Andrea Kottke,"

Albrecht Wendel^l

We examined leukocyte counts and ex vivo cytokine response of whole blood to lipopolysac- charide (LPS) or lipoteichoic acid (LTA) in patients under low-dose molgramostim therapy.

Patients were injected subcutaneously daily for ten days with 1 flglkg(n=9)or 2 flglkg (n=14) molgramostim. Leukocytosis was observed in all patients, but only the eosinophil fraction was significantly increased in relation to other leukocyte populations. Ex vivo IFN-y release was decreased and IL-IO and IL-lra secretion were increased in response to LPS or LTA.

Thus, in non-neutropenic patients, leukocytosis can already be initiated by low doses of molgramostim. The ex vivo cytokine data suggest that these doses prime blood towards a systemic anti-inflammatory response.

The pluripotent haematopoietic growth factor granulocyte-macrophage colony-stimulating factor (GM-CSF) regulates proliferation, differentiation and mature functions of granulocytes and monocytes/

macrophages. Recombinant human GM-CSF (rhGM- CSF) shortens idiopathic leukopenias and reduces risk of infection.¹This protection may also result from the stimulation of mature immune cell functions, an effect which, if not tightly controlled, can exacerbate the inflammatory reaction which may lead to tissue damage and septic shock. Various in vitro, ex vivo and in vivo experiments have shown that rhGM-CSF potentiates the release of the pro-inflammatory cytokines TNF-n and IL-l in response to LPS.^I^:-³

Though administration of rhGM-CSF (750 ug/nr' per day i.v.) to sarcoma patients with neutropenia increased ex vivo LPS-stimulated TNF-n and IL-l~

From the 'Department of Biochemical Pharmacology, University of Konstanz, Konstanz; 2Department of Dermatology, University of Bochum, Bochum; 3Essex Pharma GmbH, Munich, Germany Correspondence to: Thomas Hartung, University of Konstanz, Biochemical Pharmacology, 78457 Konstanz, Germany. E-mail:

Thomas.Hartung@uni-konstanz.de

Received 27 August 1999; received in revised form 22 May 2000;

accepted for publication 30 May 2000

1043-4666/00/101570+05 $35.00/0

KEY WORDS: blood/cytokines/growth factors/human/inflammation

1570

release by monocytes, this high dose administered daily for two weeks did not affect the spontaneous release of these cytokines." In another study in cancer patients, one single dose (2.5, 5, or 10 ug/kg) of rhGM-CSF resulted in a significant increase of plasma levels of the anti-inflammatory mediator IL-l receptor antagonist (IL-lra) and a trend towards increased levels of the chemokine IL-8.⁵ These observations combined with the relatively low incidence of inflammatory side effects reported in the several hundred thousand patients already treated with rhGM-CSF, led to the hypothesis that lower doses of rhGM-CSF might sufficeto initiate leukocytosis without exacerbating inflammation. In the context of a dose-finding study examining the potential of rhGM-CSF in wound healing," we tested this idea.

Whole blood from rhGM-CSF treated patients was exposed to the bacterial components lipopolysac- charide (LPS) or lipoteichoic acid (LTA) ex vivo, so allowing the assessment of alterations of the humoral immune reaction without exposing patients to these stimuli."

RESULTS

Tolerability

In the group receiving 2 /lg/kgrhGM-CSF, 27 cases of side effects were recorded while there were only

CYTOKINE, Vol. 12, No. 10 (October), 2000: pp 1570-1574 Konstanzer Online-Publikations-System (KOPS)

URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-opus-79388 URL: http://kops.ub.uni-konstanz.de/volltexte/2009/7938/

(2)

'3..... 10000

Q)

P- I,"

.$ ⁱ

"ai

'"

'"8₀ ^....^~

:E

...

Q)

:.a~

-a

5000

...

~

o

3'

Day of treatment

1571

Figure1. Leukocyte counts under molgramostim treatment.

Changes in.leukocyte counts bydaily subcutaneous injections ofmolgramostimin a group of nine patients receiving a dose of lug/kg (leftbars) and ina group ofl d.patients.receiving21!g/kg (right bars),respectively.Blood sampleswere taken shortly beforethe next injection.Neutrophils,

o

eosinophil s,~basophils,. ;monocytes,~Iymphocytes,f:iiI.

TABLE 1. AbsoliiWand_relati_~eerlsinophlInillnbers_dU~ihgt~e courSeofmolgra~ostimtr~tnlent

day I day2 " ,d ay) dayS ^" day IQ

n= 23 14 IS. 1J 7

eosinophils/ul . 101 ±,L8 369 ±48 352±42 421 ± 76 890 ± 140

eosinophil fraction 1.6±'a.2% 3.0 ± 0.4% 3.4± 0.3% 4.6±0.6% 7.8 ± 1.1%

Responses of patients treated witheithi,r'lug/kgor 2~gJkgmolgramostim perday<M not differsignificantly.fhereforethe data from

,both groups werecombined.Dataaremean:±; SEM,

.1-

13 in the .l.ug/kggroup. No,severe adverse events occurred." ,... ' " . "

Leukocyte,collnts'; ,,

In'both dosage groups, the .Ieukocyte.:count doubled,by,day 2 and remained elevated above-basal values (Fig.i: l). .These..changes, mainly-.reflected increases,in neutrophil-and monocyte counts. .Also, the eosinophilⁱcount rose steadily,:reaching nine-fold initial values

On

dayIO: (Table 1)., _

: <:.' . '. ~",

Cytokine release, " '

Spontaneous release of TNF-dand'IL-8iilcreased slightly on day 2 (Fig. 2A'and ~), but was slightly decreased when calculated per monocyte(Fig;2D).

Spontaneous IFN-y release was unchanged (Fig.,2.B):

, , In, patients' blood .stimulatediwith LPS, we obser-v'ed. ^- ^-an^" initiaLincreas~ln::INF~arelease^'^" . ^". ^" ' : ;" ^foilowed

by,,a decrease after prolongedjreatment,(Fig. 3A);

Intracellular.TNF-n.measurementswith fluorescence- labelled anti-TNF-a antibodies.vin ol.Ps-stimulated blood detected only monocytes as producers ofTNF-a. (data not shown). So, we calculated the release of TNF-a per million monocytes and found.rthisr.was unaltered throughout the study (Fig. 3B), indicating that the.totalincrease-in TNF~aproduction stemmed from'normal,..competent.andnotpre-activatedcells.

There'wasa trend'towards a decrease-in total IFN~y

release (Fig.3C) and the productionof'Hxl~(Fig: 3D) onlyincreasedafter prolongedrh6M~CSFtreatment;

while'IL~8'or,IL-12production were hardly affected (Fig;,3E'andF).

(3)

1

," .

10

v,

4 6 10

Day ofst u dy

o

E

~

0.1

~

tiJ ⁴

i

_d ³ ^D

S0 cc0

...

...~..

E

~

'~" ~

0 2

10 4 6 8 Day of study 2

.3.--- -- - - ---,

~

0.5 E

~tl .Eo<Z

0

Figure 2. Spontaneouscytokine release.

Spontaneous releaseofcytokinesby'whole blood from patients treated with lug/kg-or-zug/kg molgramostim per day. The cytokine concentration was determined by ELlSA after 24 h of incubation/Thereleaseof TNF-a was also relatedto monocytes iri the incubation. A, TNF·a;B,IFN ..t:C, IL-8; D,TNF-a permon~yte. · ~} l'-;.~ ·

,:.', ~ '.~^.

DISCUSSION '/ "

Although the molgramostim doses used were well below'the..recommended dosage-for cancer"patients (5-10 ug/kg'per day),"a substantialIeukocytosis-was recorded.The;disproportionate.increase of theeosine- phil.'.count :may be undesirable':,Accumulation.of eosinophilsandmacrophages .afteradeneviral:expres- sion of GIvI-CSF in rat lung was associated.withtissue

LPS-stimulat.~dexvivo.release9qL~lra(Fig.4A) .' ..inj ury.•and fibrosis,S,Ho.'!Vever, jn a"p atient with quadrupled by day 2 and remained elevated, G-CSF T-lymphocytosis,granulocyt6penia:and -severe peri- andIL~lOrelease (Fig. 4B and C) increased transiently, anal infection, eosinophilia initiated by rhGM-CSF reflecting changesin monocyte counts. The secretion of treatment correlated with-improvementof the perianal IL-6persisted significantly above basal levels(data not ulceration.⁹ " ,

shown). ',' . " Other stu<.liesonCSF treatmerit1havemeanwhile The lower,dose of I1101grliwostim produced a .. sh,gwnthatthech()sen limit 0[15,'090leukocytesperIII stronger trend toward an anti-inflammatory.profileof blood, was somewhat conservative. We also found cytokine release on stimulation: here, the concen- "that thenew'Ieukocyteswerenotpre-activated. One ug tration of IL-lOreleased was always greater and the and 2 ug/kg molgramostim pet: day were equipotent amount of IFN-y was always less,. in .stimulating haematopoiesis while the l ug/kg , The'cytokine rresponse to LTA'corresponded group reported fewer side effe~tsLlndicating tharlow closely withthatdescribed:for·LPS.The amounts qf doses of rhGM-CSF might be.more tolerable in non- TNF-u;'G-CSFandIL~lO r~lea:sed"weresimilaT with neutropenic patients while retaining desired effects.

both stimuli,while lessIFN-'y'and more i:c.:8arid IL-6 The cytokine response pat!efutowaidsinflamma- wereinducedby LTA. " . . , ^;f tory stimuli'offered unexpecte~rfesults: lowdoses of molgramostimpromoted an arid-tnflanlluatory profile, primingimmune cells_:to produC~;1nore IL-IO,G-GSF and-Ils-Irarwhile release

of

prg:iriflammatoryTNF-u, IFN- and IL"'ll3' was not affecte~i9tslightlY:decreased.

The distinct increasein Ilf6release-maybe-considered a protective reaction that is initiated sooner under molgramostim treatment.!" Notably; oytokinerelease onday2wasmot;diffetentinpatientswho-dropped out of or who completed treatment.

.'GM"CSF ;was previously classed a.;pro- inflammatory',cytokine.".Now .it .appears-that 'the

(4)

573

150,---,--~---~-__,_,

10 4 6 8 Day of study 2

B

o

50 100

:2 60 ~

<ll

..;< ^Q)

{:'lx ~₀

-e0 40 §

~ '"^El

~

20 ⁰^,..;

E

Ees es

rd, _~

Z Z

Eo< Eo<

0 2 4 6 8 10

Day of study

10

**

4 6 8 Day of study

:2 ¹²⁰ _D

..;<

C'l

x 90

"d

..s

0

..c 60

I

₌ ³⁰

,..;

...~

0 2

4 6 8 10

Day ofstudy 2

:2 1.5 _F

..;<

C'l X

-e₀ 1.0

..s

..c

I

C'l ^0.5 ,..;

...~

0 2 4 6 8 10

Day of study

o

4

:2 "E

..;<

C'l_x 3

'g

:a

2

~

E 1

<:J:)

~

Figure 3. Pro"infta~atory~ytokine.pattern.

;' .... j

Release .of pro-inflammatory .cytokines by whole••blood. of patients who received I ug/kgor 2 /lglkg molgramostimper. day, incubated With 10/lg!ml LPS for24 h. ThereleaseofTbllt-« calculated permonocyte is also shown (B).'A,TNF-a;C,IFN"y;n,IL-l~;E, IL"8; F, IL-12.

immunostimulatoryproperty. ofrh.GM':CS~,'i.e.• its ability to recruit potentgJ;~nu1ocytes,and monocytes,.is already effected at low doses with relatively mild side effects, opening up new potential indications.

lll()rningso~d~ys1,},3,Sand,IQ before drug injection. Of the.group rece~ving 2 /lWkg n,}olgramostim, six patients withdrew from the study and. six patients were dismissed because their white blood cell count (WBC) rose above 15 OOO/Ill. In the I !!g/kg group, three patients were dismissed for this reason.

MATERIALS AND METHODS Study design

Twenty-three patients of both sexes aged 40 to 85 who had a non-ulcerating basalioma of the face or neck removed by cryosurgery were enrolled into the open-label study.

Patients received 2/!gIkg(1l=14) or Lug/kg (n=9) molgramostim (Leukomax'P, provided by ESSEX Pharma, Munich, Germany), subcutaneously for 10 days starting on the day of surgery. Blood samples were collected in the

Laboratory measurements

Automated differential WBC were performed. Five ml 20% whole blood in RPMIl640 (Biochrom, Berlin, Germany) supplemented with:4mM glutamine, 100 ID penicillin, 100 ug/ml streptomycin and 2.5 ID heparin (Hoffmann LaRoche, Grenzach~Whylen, Germany) was stimulated with 10 ug/ml LPS from Salmonella abortus equi or 10 /!g/ml LTA from Staphylococcus aureus (both Sigma, Deisenhofen, Germany). After24^11-at37°Cvials were shaken

(5)

REFERENCES

of Anke Biedermann, Elisabeth Schmid and Volker

Q~p~e~ J. . ,~"

^"^,

o

"--'-j~~'.

**

1 15 20

I

_o

....

~ .

....

1.Hartung ..'T (1999) Immunomodulatien .by colony-.

siimulating factors;RevPhysiolBi.QchernP'harma~1136:1"':164, , 2.Tiegs G; BarsigJ, MatibaB,UWig S, Wende1A (1994) Potentiation by granulocytemacrdplj~e'~ny-s'ffmu1atingfactor of¥popolysaccharidetoxicityin

mtee.

J Clin InveSl-93:2616:-2622. .

i 3. Bundschuh'D~,Harswj,!HartungT,R~dow^{F, Docke} .W1i>r'¥olk:H"n'.\Y:!(!L~.~(r9myranu10~¥.te-lIl\ilrophagecolony- stimulatingfactor and'ipN-grestore the systemicT~~F-aresponse to-. endotoxin vin"Iipopoljsaccharide-descnsitized mite. J Immunol 158:2862-287 1.. ••. .•.... . ' .. " .

:2

... _'_.'L P~r](jns, ~(;,..Yi!.Q!,J..ll,n: R at ,SLScneule R'&, H amilton R, Helian Ns(1993)_L_Eff~cts¥of'conlinuo us l}igh dose rhGM-CSF infusion_{5, Arnan}⁰^,,.n.hUma~M:t'S'focl(dreher K , Thews A,.^" ^v^.ⁿ^.~9'Je:

. .

ac.tivity.Am..'sJ.He~atoI4.3:2Kienast K, Aulitzky.79.-~85.

'YE, Faiber L, HausU, KociB,Huber C, Pesche1 C (1996) :R eglllatiqn- of-iminun·omodulatorY"'fuI\(itions .~y. granulocyte-

i.in.,.a..cro.p~~g.e COlo.~y-.s~im. ^ul^.•^.^at.i^.n..^g ^t:^.a^.^c^.tor:.!ia¥ ^g.ratru^.:..ulocyte colony- .stimulatih g factor mVIVO.Ann,Hematoln~31-2~.

\ •. 6.'~offmann K,Freita,~.-M!;" S;iic~er"M;:',Hoxtermann S,

;UllrichS:.Altm eyer P.cl~9S(Phase·i1l¥"D ose-rariging"Studie zur

!u~fwk~ung.~er"T(Jrera~ilitat von .rhq~5cSFg§l'v101~ram~stim)

isowied~SJienEmfluB aufdie Wundheilungtfiach.Kryochirurgiedes

•Basalzellkarzinomas, AktDermatoI24:67~74. )} .

: 7. '~ernoff.AE, Granowitz BV" $4apirO§L, Vannier E,

,Lonnemap,11 G, Angel JB,.KennedY"JS, R.absons-AR,_Wolff SM,

!Dinarello,;tA (1995).A·.randomized, controlled!t;ial-of' IL-1 0 in

!humans;.;iinhibitioJ'l,;oLi nflammaio ry.•cytokine production and im:HJ:unet~§ponses, (gImmuaol 154~5492-:-54919.

·P~M$.lf!~t;1.!<i.Y ^{8. JQ.}^hg~v~il:W,WY' Jordana M, Graham F, Gau1dieJ

. . . . . . (I996) 1ft~nsfer of granulocyte-macrophage colon~-stimulatin~

Figure4. Antt-mftammatory cytokme pattern. factor gen~ to rat lufi1i'!}ifi{lu~§l(\!(Hi/loIlJTtIlli:~\¥i~1'i'dgytosis';I'a1id

" .. . fibrotic rt1a~tions.J Clin Invest 97:1102-1110.

Gour~e^>1q,f~J;,PS':'A9Ue;tblll.n"~~t,,mtlil:~j~~^rn:edW~~-,~!e~4· ',"'1':, ,;:"-;l'91'~ieget'ia:cKl'{eHiiM~'ivelifueyet!'K!;"Nigi:i1(5'A"W elte'K

by:}VIW1ei1ll0.0d.~pfiip~Ie)!ts.,tteat!l9:~~Itb.-11; J.lgl,kggor,;q2·IU¥!%,","~1(1990rU~)Cf're&Jiill;lnarlt?h\iihan"grllfiu18dyteliriacrop1i&te'CBlbny~

molgramostirnper day,A,IL-1ra; H, G-CSF;C, IL-lO, stimulatin1tfactorinTvlymphccytosiswith granulocytopenia.EurJ Haematol44:205--206: .

1O.){ingZ,GauldieJ,CoxG;BaumannH,JordanaM,Lei

~,'....~hq~'~·IM(!1l?~)-HL-(;1ri~,<jtIJJl'lln*'!'lffl,!.@l,lfi~Qr¥r)~Y.tlillQW;

~1q.~~~);qfefRQI,SW1f~,gl-{lt·11?11cW,1·t,:1BI>''i,~.nrml9q~':W~~:j·}~~fRg

relipol'\s~~;1.~n nvrsl' :5 I-j LU..~ ., ' '., . •