jak inhibition impairs nk cell function in ...microenvironment and immunology jak inhibition impairs...

Microenvironment and Immunology

JAK Inhibition Impairs NK Cell Function inMyeloproliferative NeoplasmsKathrin Sch€onberg1, Janna Rudolph1, Maria Vonnahme1, Sowmya ParampalliYajnanarayana1, Isabelle Cornez1, Maryam Hejazi2, Angela R. Manser2,Markus Uhrberg2,Walter Verbeek3, Steffen Koschmieder4, Tim H. Br€ummendorf4,Peter Brossart1, Annkristin Heine1, and Dominik Wolf1

Abstract

Ruxolitinib is a small-molecule inhibitor of the JAK kinases,which has been approved for the treatment of myelofibrosis, arare myeloproliferative neoplasm (MPN), but clinical trialsare also being conducted in inflammatory-driven solidtumors. Increased infection rates have been reported in rux-olitinib-treated patients, and natural killer (NK) cells areimmune effector cells known to eliminate both virus-infectedand malignant cells. On this basis, we sought to comparethe effects of JAK inhibition on human NK cells in a cohortof 28 MPN patients with or without ruxolitinib treatmentand 24 healthy individuals. NK cell analyses included cellfrequency, receptor expression, proliferation, immune syn-apse formation, and cytokine signaling. We found a reductionin NK cell numbers in ruxolitinib-treated patients that was

linked to the appearance of clinically relevant infections. Thisreduction was likely due to impaired maturation of NKcells, as reflected by an increased ratio in immature to matureNK cells. Notably, the endogenous functional defect of NKcells in MPN was further aggravated by ruxolitinib treat-ment. In vitro data paralleled these in vivo results, showinga reduction in cytokine-induced NK cell activation. Further,reduced killing activity was associated with an impairedcapacity to form lytic synapses with NK target cells. Takentogether, our findings offer compelling evidence that ruxo-litinib impairs NK cell function in MPN patients, offering anexplanation for increased infection rates and possible long-term side effects associated with ruxolitinib treatment. CancerRes; 75(11); 2187–99. �2015 AACR.

IntroductionMyeloproliferative neoplasms (MPN) are clonal bone mar-

row stem cell disorders in which the proliferation of an abnor-mal clone of hematopoietic progenitor cells in the bone mar-row results in hypercellular bone marrow, but may also lead tosevere fibrosis in myelofibrosis (1). Janus kinases (JAK) aretyrosine kinases playing an important role in the transductionof cytokine signals (2). Almost all of the polycythemia vera andapproximately half of the myelofibrosis patients harbor a gain-of-function mutation that results in a V617F amino acid change

in the JAK2 protein, mediating constitutive activation of theJAK/STAT (signal transducer and activator of transcription)pathway (3). Ruxolitinib is an oral JAK inhibitor alreadyapproved for the treatment of myelofibrosis and polycythemiavera. Although ruxolitinib is supposed to be not curative, itleads to reduction of increased blood counts and an excellentsymptom control linked to a substantial reduction of proin-flammatory mediators and reduction of spleen size (4, 5). Theobservation that the therapeutic effects are irrespective of thepatients' JAK mutational status and that the compound inducesonly limited anticlonal activity (6) suggests that it profoundlymodifies the inflammatory microenvironment. The idea thatJAK inhibitors are immunosuppressive is underscored by anincreased infection rate of patients treated with ruxolitinib(7–11), but also by its potential benefit in inflammatory-drivencancers, such as pancreatic cancer with increased C-reactiveprotein levels (12). In line with these clinical observations, werecently provided evidence that JAK inhibitors markedly impairdendritic cell biology (13).

Natural killer (NK) cells are another innate immune effectorcell population recognizing and killing malignant or virus-infected cells (14). NK cell function has to be tightly regulatedby a complex balance between various activating and inhibitoryNK cell receptors (15). Moreover, cytokine signals mediatedvia the JAK/STAT pathway are key determinants for NK cellactivation (16).

The role of NK cells in Philadelphia-negative MPNs ispoorly understood. Thus, the aim of this article was to

1Medical Clinic, Oncology, Hematology and Rheumatology, UniversityClinic Bonn (UKB), Bonn, Germany. 2Institute for TransplantationDiagnostics and Cell Therapeutics, University Clinic D€usseldorf,D€usseldorf, Germany. 3ZAHO, Bonn, Germany. 4Department of Hema-tology, Oncology, Hemostaseology, and Stem Cell Transplantation,Faculty of Medicine, RWTH Aachen University, Aachen, Germany.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

K. Sch€onberg and J. Rudolph contributed equally to this article.

Corresponding Author: Dominik Wolf, Medical Clinic, Oncology, Hematologyand Rheumatology, University Clinic Bonn, Sigmund-Freud-Strasse 25, 53127Bonn, Germany. Phone: 49-228-287-17233; Fax: 49-228-287-51729; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-14-3198

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2187

on June 20, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst April 1, 2015; DOI: 10.1158/0008-5472.CAN-14-3198

http://cancerres.aacrjournals.org/

characterize NK cells in MPN patients and to define theimpact of JAK inhibition on NK cell activation. The data mayhelp to better understand therapeutic but also potential sideeffects of JAK inhibitors in MPN as well as in other cancerpatients.

Materials and MethodsPatient samples

Blood was taken from 28 MPN patients and 24 healthydonors. Patient characteristics are shown in Table 1. Relevantinfections were defined according to the CTCAE4.0 gradingsystem as grade 2 (i.e., requiring systemic treatment withantivirals, antibiotics, and/or antifungals). Our study wasapproved by the local ethical committees (No. 154/13 forBonn; EK127/12 for Aachen; No. 3462 for D€usseldorf) andwas performed according to the declaration of Helsinki. Buffycoats were obtained from healthy blood donors (UniversityHospital Bonn, Bonn, Germany). NK cells were isolated bymagnetic bead separation (NK Cell Enrichment Kit; StemCellTechnologies), and purity (>95%) was routinely checked byflow cytometry.

Cell cultureThe K562 target cells were cultured in RPMI-1640 with 1%

penicillin/streptomycin (both from Life Technologies) and 10%FBS (Biochrom). K562 (kindly provided by Bettina Langhans,University Clinic Bonn, 2012) were regularly tested forMHC classI absence. The NK cell line NK-92 (kindly provided by HelmutSalih, University Clinic T€ubingen, Germany, 2012) was culturedin Eagle's Minimum Essential Medium alpha (MEM-a) contain-ing 20% FBS, 1% penicillin/streptomycin, and 100 U/mL recom-binant IL2 (Proleukin, Novartis). Expression of NK markers aswell as functional capacity was regularly tested by flow cytometry.All cell lines were tested for Mycoplasma monthly (MycoAlert,Mycoplasma detection Kit Lonza).

Functional NK cell assaysFor in vitro experiments, freshly isolated NK cells were

cultivated overnight with 1000 U/mL IL2 and increasingconcentrations of ruxolitinib (dissolved in DMSO, Selleck-chem, Houston, TX). Alternatively, NK cells were activated viaNKp46 (Biolegend) cross-linked by plate-bound goat anti-mouse IgG (Dianova). Peripheral blood mononuclear cells(PBMC) or NK cells isolated from MPN patients and healthycontrols were used for ex vivo functional analysis. Functionalassays were performed as described before (17, 18). Briefly,NK cells were cocultured for 4 to 6 hours with K562 at theindicated ratios. IFNg production was analyzed by intracel-lular cytokine staining and subsequent flow cytometry 6 hoursafter stimulation with IL12/IL18 (Immunotools/Biozol) andBrefeldin A (Sigma-Aldrich). For synapse formation assays,CellTrace Violet-labeled NK-92 and CellTrace Far Red–stainedK562 cells (Life Technologies) were cocultured at the indi-cated effector-to-target ratios.

Flow cytometryFor flow cytometry, CD3-FITC, CD3-APC, CD56-PE, CD56-PE-

Vio770, NKG2D-PE, NKp46-APC, CD16-APC-Vio770, CD158a/h-Vioblue, CD158b-APC, CD158e-PerCP (all from Miltenyi Bio-tec), CD107a-FITC,CD69-PerCP-Cy5.5, CD4-PerCP-Cy5.5, CD8-APC-Alexa750, CD27-PECy7, CD161-APC, CD57-APC, CD11b-Bv421 (Biolegend), andCD159-PE (BeckmanCoulter)wereused.Intracellular staining was performed using IFNg-FITC, GranzymeB-FITC, Perforin-Vioblue (Miltenyi Biotec) with Cytofix/Cyto-perm (Biolegend).

Phosphoflow cytometry was performed as described previ-ously (19). In brief, IL2-stimulated NK cells exposed toincreasing concentrations of ruxolitinib were fixed with PFAand subsequently barcoded with Pacific Blue SuccinimidylEster (Life Technologies). Phosphorylation of targeted pro-teins such as S6, STAT5, and ERK (BD Biosciences and CellSignaling Technology) were analyzed. All flow cytometry

Table 1. Patient characteristics

Healthydonors

Age-matchedhealthydonors

MPN withoutruxolitinib

MPN withruxolitinib

Number 12 (þ10 BC) 12 12 16Median age (range) 30 (24–40) 71 (69–87) 61 (47–78) 61 (30–86)Gender (female/male) 6/6 4/8 6/6 8/8Type of MPNPMF — — 10 9Post-PV MF — — — 3Post-ET MF — — — 1PV — — 2 3

DIPSS in MF before ruxolitinibLow risk — — 3 1Int-1 — — 4 4Int-2 — — 2 5High risk — — 1 3N/A — — 2 3

Median duration of ruxolitinib (mo) — — — 10.5Relevant infections n.d. n.d. 2 9Viral n.d. n.d. 1 6Bacterial n.d. n.d. 1 2Fungal n.d. n.d. 0 1

Abbreviations: BC, buffy coat; DIPSS, Dynamic International Prognostic Scoring System; MF, myelofibrosis; PMF, primary myelofibrosis; PV, polycythemia vera; ET,essential thrombocythemia; N/A, not applicable; n.d., not determined.

Sch€onberg et al.

Cancer Res; 75(11) June 1, 2015 Cancer Research2188




analyses were performed on a FACSCanto II (BD Biosciences)and analyzed using either Diva Software, FlowJo, or www.cytobank.org.

Proliferation assayFreshly isolated NK cells or PBMCs were stained with

0.5 mmol/L CellTrace CFSE (Life Technologies) and sub-sequently cultivated for 5 days in RPMI in the presence of200 to 1000 U/mL of IL2. Proliferation was detected by car-boxyfluorescein diacetate succinimidyl ester (CFSE) dilutionand analyzed by FACS.

Confocal microscopyNK-92 and Far Red–labeled K562 cells were cocultured

for 30 minutes in the presence of 1000 U/mL IL2 and allow-ed to settle on PLL-precoated (Sigma-Aldrich) glass cover-slips. For immunofluorescent staining, cells were fixedusing 4% formaldehyde and permeabilized with 0.2% TritonX-100/PBS. Unspecific binding of antibodies was blockedwith 3% BSA/PBS. Perforin was stained with the primaryantibody (Thermo Fisher Scientific; clone delta G9, 1:100in PBS) overnight at 4�C, washed and incubated for 1 hourwith a goat anti-mouse AlexaFluor405–coupled secondaryantibody (Life Technologies; 1:200 in PBS), together with

Figure 1.Ruxolitinib (ruxo) affects NK cellfrequency and number in patientswithmyelofibrosis. A, representative dotplots of the flow cytometry analysis ofa healthy donor, an age-matchedhealthy donor, a myelofibrosis patientwithout treatment, and amyelofibrosis patient treated withruxolitinib are shown. B, top,frequency and absolute NK cellnumbers were calculated from flowcytometry analysis and are shown asscatter plot from healthy donors(n ¼ 12), age-matched healthy donors(n ¼ 12), MPN patients withoutruxolitinib treatment (n ¼ 12), andpatients receiving ruxolitinib therapy(n ¼ 15). Individual patients and themean are shown for each group(� , P < 0.05; �� , P < 0.01). Bottom, thepercentage of CD56þCD3� NK cellsand the absolute NK cell numberplotted against the duration ofruxolitinib administration in months.The line represents the linearregression with 95% confidenceintervals (n ¼ 27). C, frequency andabsolute NK cell numbers of MPNpatients while on ruxolitinib treatmentdivided into patients without and withrelevant (CTCAE �grade 2) infections(n ¼ 16). The scatter plots provideindividual values and the mean(� , P < 0.05; �� , P < 0.01).D, representative dot plots of the NKcell percentage prior to and during(left and middle plots) and 3 monthsafter stopping ruxolitinib (right plots)are shown. E, CD56bright:CD56dim ratiowas calculated from healthy donors(n ¼ 12), age-matched healthy donors(n¼ 12), MPN patients (n¼ 12) withoutruxolitinib, and patients withruxolitinib treatment (n ¼ 15) by flowcytometry. Bars, mean � SEM.

Ruxolitinib Impairs NK Cell Function

www.aacrjournals.org Cancer Res; 75(11) June 1, 2015 2189




Figure 2.MPNpatients have an endogenous defect of NK cell effector functions, which is further aggravated by JAK inhibition. A, freshly isolated PBMCswere coculturedwithK562 for 4 hours in the presence of 1000 U/mL IL2. Degranulation was detected by flow cytometry analysis of CD107 expression in a 1:1 ratio for healthydonors (n¼ 13), MPN without ruxolitinib (ruxo; n¼ 10), and MPN with ruxolitinib (n¼ 14; �� , P < 0.01; �� , P < 0.001). For analysis of NK-mediated target cell killing,CFSE-stained (0.5 mmol/L; Life Technologies) K562 and NK cells from healthy donors (n ¼ 13), MPN without ruxolitinib (n ¼ 11), and MPN with ruxolitinib(n ¼ 16; �� , P < 0.01; �� , P < 0.001) were cocultured for 4 hours in a 10:1 effector-to-target ratio in the presence of K562. (Continued on the following page.)

Sch€onberg et al.





Phalloidin-TRITC (50 mg/mL; Sigma Aldrich) for analyz-ing filamentous actin. The slides were examined at roomtemperature using the microscope and camera model Olym-pus Fluoview 1000 Confocal LSM with PlanS Apo60�/NA1.35, oil immersion. Images were processed with AdobePhotoshop.

Statistical analysisStatistical analyses were performed using GraphPad Prism

Software. If more than two groups were compared, ANOVA withsubsequently the Tukeymultiple comparison test was performed,whereas when two groups were compared, the Mann–Whitney Utest was applied. Values for P less than 0.05 were consideredstatistically significant. Correlation was calculated with the Pear-son/Spearman test.

ResultsRuxolitinib reduces NK cell levels in MPN patients

First, NK cell frequency from patients with myelofibrosis orpolycythemia vera was analyzed by flow cytometry (fordetailed patient characteristics, see Table 1). In ruxolitinib-na€�ve patients, mean NK cell frequency (12.63% �1.81%) wascomparable with young (13.51% �1.44%) as well as to age-matched healthy donors (12.58% �1.77%). In contrast,patients treated with ruxolitinib have clearly reduced levelsof NK cells (5.47% �1.27%; Fig. 1A and B, top). The differ-ence between absolute NK cell numbers of MPN patientswith and without ruxolitinib is not highly significant(P < 0.01), due to the fact that two patients received ruxolitinibonly for 2 months. We also provide clear evidence thatthe reduction of NK cells develops over time (Fig. 1B, bottom;P < 0.001 for relative and absolute NK cell quantification). Ifwe focus only on patients having received ruxolitinib for atleast 3 months, the difference between MPNs with and with-out the drug was highly statistically significant (P ¼ 0.0005;data not shown).

To evaluate a potential clinical implication, patients underruxolitinib were classified in two groups: Patients with andwithout relevant infections (i.e., grade �2) according toCTCAE4.0. Interestingly, 56%of our patients receiving ruxolitinibdeveloped relevant infections, whereas only 16% of patientswithout ruxolitinib treatment did so (see Table 1). More impor-tantly, the NK cell frequency is linked to the appearance ofinfections, as patients with relevant infections during ruxolitinibtherapy had a significantly lower NK cell percentage and absoluteNK cell counts than those patients without relevant infections(Fig. 1C).Of the total infected patients with ruxolitinib treatment,approximately two third (66%) experienced viral infections(including herpes zoster).

To follow the time-dependent impact of ruxolitinib withinindividual patients, two patients (MPN 8 and 9) could be sequen-tially analyzed, and the drop of NK cells during ruxolitinib after 3months could be clearly seen. Another patient (MPN 4) had tostop ruxolitinib therapy due to an adverse event. In this individ-ual, NK cell levels rose back to normal values, indicating that theNK cell drop was ruxolitinib dependent and is potentially revers-ible (Fig. 1D).

We next evaluated whether ruxolitinib affects the CD56dim

and CD56bright NK cell distribution (mirroring differentfunctional characteristics: CD56bright predominantly producecytokines, and CD56dim NK cells are cytotoxic). In bothhealthy donor groups (young and age-matched), the ratioof CD56bright:CD56dim NK cells was approximately 0.1. InMPN patients without ruxolitinib therapy, the CD56bright:CD56dim ratio is increased, as the frequency of CD56bright

NK cells is almost doubled when compared with healthy(age-matched) donors. The ratio in ruxolitinib-treatedpatients is comparable with the level seen in healthy age-matched controls.

MPN patients have an endogenous defect in NK cell functionthat is further aggravated by JAK inhibition

The function of NK cells in MPN patients was analyzed byCD107 expression and a classical killing assay. Degranulationand killing were significantly lower in MPN patients than inhealthy controls (Fig. 2A). In healthy donors, no strong age-dependent reduction in the functional capacity of NK cells canbe observed (20). If we compare NK cells from healthy donorsto MPN patients with and without ruxolitinib, the functionalactivity of NK cells isolated from MPN patients is clearlyreduced (Fig. 2A). The use of highly purified NK cells, insteadof PBMCs, showed that the difference between the patientswith/without ruxolitinib is probably mainly a consequence ofthe difference in NK cell numbers. The few remaining NK cellsisolated from ruxolitinib-exposed patients show only a slightfunctional impairment when compared with NK cells fromruxolitinib-na€�ve patients (Fig. 2B).

We next phenotypically characterized NK cells from ruxoli-tinib-na€�ve versus ruxolitinib-exposed patients and comparedthem with normal NK cells from young as well as from age-matched healthy donors. Ruxolitinib-treated patients had aclear reduction of the activation marker granzyme B whencompared with MPN patients without treatment and bothhealthy control groups (Fig. 2C). In aged donors (i.e., age-matched controls and untreated MPN patients), the number ofNKp46-, NKG2D-, and perforin-expressing NK cells wasreduced. Interestingly, treatment with ruxolitinib leads to afurther reduction of the activating receptors NKp46 andNKG2D even when compared with age-matched controls(Fig. 2C). NKp46 and NKG2D expression in CD56dim and

(Continued.) Killing activity was evaluated by quantifying cells that were double positive for CFSE and propidium iodide. B, highly purified NK cells of MPNpatients without ruxolitinib (n ¼ 10) and MPN patients with ruxolitinib (n ¼ 10) were cultured and analyzed as in A (�� , P < 0.01; �� , P < 0.001). C,receptor frequency of five different NK cell receptors was determined on gated NK cells from healthy donors (n ¼ 22), age-matched healthy donors (n ¼ 12),patients with myelofibrosis without ruxolitinib (n ¼ 12), and patients with myelofibrosis (n ¼ 16) receiving ruxolitinib treatment. The individual patientdata are shown, as well as a line representing the mean (� , P < 0.05; �� , P < 0.01; �� , P < 0.001). D, representative histograms show proliferationassays with PBMCs from a healthy donor and MPN patients without or with ruxolitinib. PBMCs stained with 0.5 mmol/L CFSE were cultured for 5 days with IL2.Histograms show CFSE dilution after gating on NK cells (CD56þCD3�).






Sch€onberg et al.





CD56bright NK cells was not statistically different (data notshown). Similar to reduced killing activity of primary NK cellsfrom MPN patients, proliferation was reduced when comparedwith healthy controls, even though analysis of NK cell prolif-eration was technically challenging due to the very low numberof NK cells (Fig. 2D).

Reduced NK cell frequency in ruxolitinib-treated MPN patientsis caused by impaired NK cell maturation

The majority of NK cells in healthy donors are doublepositive for NKp46 and NKG2D. Although age-matched con-trols already exhibit reduced levels of double-positive NKcells, the population is further reduced in ruxolitinib-treatedpatients, whereas double-negative NK cells are increased(Fig. 3A). Double-negative cells may reflect an increased fre-quency of probably functionally immature NK cells. Therefore,we analyzed NK cell maturation stages in more detail, bystaining for CD27/CD11b expression, allowing the definitionof the different NK cell maturation stages (21, 22). The fre-quency of mature NK cells (CD11bþCD27�) is slightly reducedin MPN patients without ruxolitinib but is significantly lower inpatients with ruxolitinib treatment than in age-matchedhealthy donors (Fig. 3B). Next, we used KIR and NKG2Aexpression to classify NK cells into four maturation stages(23). Strikingly, only MPN patients exposed to ruxolitinib hada significantly higher frequency of NK cells lacking expressionof NKG2A and KIR (NKG2A�KIR�; Fig. 3C). The NKG2A�KIR�

subset represents an immature NK cell differentiation stagelinked to functional hyporesponsiveness (24). In summary,ruxolitinib seems to induce a NK maturation block, reflectedby the increased frequency of immature and decreased abun-dance of mature NK cells.

The reduced frequency of NKp46þNKG2Dþ NK cells may,at least in part, be due to increased expression levels of theirligands on the MPN clone. As it is difficult to specifically gateon the MPN clone by flow cytometry (due to a lack of anunambiguous marker), we analyzed NKp46 and NKG2Dligands in patients and healthy controls using fusion proteinbinding to gated CD33þ myeloid cells. Using this approach,we were not able to detect an altered expression of the ligandsin MPN patients with and without ruxolitinib therapy. How-ever, compared with CD33þ-gated cells from healthy con-trols, MPN patients expressed higher levels of NKG2D ligandin the myeloid compartment, whereas NKp46 expression wasonly marginally increased (data not shown).

Ruxolitinib impairs NK cell function in vitroBecause the analysis of NK cells from MPN patients is

limited due to low NK cell numbers, we next evaluated theimpact of ruxolitinib on NK cells in vitro. Therefore, we useddrug concentrations ranging from 0.1 to 10 mmol/L, which arecomparable with the in vivo peak concentrations achieved in

humans (1–2 mmol/L; ref. 25). Ruxolitinib dose dependentlyreduced NK cell killing activity, which was paralleled by areduced CD107mobilization (Fig. 4A, top). Stimulation of NKcells with IL12/IL18 to induce IFNg secretion was also dosedependently reduced by JAK inhibition (Fig. 4A, bottom).In contrast, JAK-independent activation via cross-linkedNKp46 was not affected by ruxolitinib (Fig. 4A, bottom). Inline with the patient's data, where we could detect furthersuppression of NK cell function by ruxolitinib, we provideclear evidence that cytokine production and killing of targetcells are dose dependently impaired by ruxolitinib. Similar toreduced degranulation and cytokine production, proliferationof NK cells is also dose dependently inhibited by ruxolitinib(Fig. 4B).

Ruxolitinib prevents upregulation of activation markersA detailed phenotypic analysis of NK cells was performed

to determine the effect of ruxolitinib on the NK cell acti-vation process. Interestingly, in contrast with the decreasedfrequency in patients NK cells, our in vitro findings using NKcells from healthy controls show a predominant effect ofruxolitinib on the expression levels of activating NK cellmarkers [reflected by mean fluorescence intensity (MFI)reduction]. The known activation-dependent shift of NKcells to the CD56bright population is inhibited by ruxolitinib(Fig. 4C). Similarly, JAK inhibition prevented cytokine-induced upregulation of CD16, granzyme B, as well asinduction of the activating NK receptors NKp46, NKG2D,and CD69 (Fig. 4C).

Moreover, as we showed that after ruxolitinib removal, thereduced NK cell frequency in patients is potentially reversible,we next investigated the potential reversibility of the ruxoli-tinib effects on healthy NK cells, or if restimulated NK cellsafter ruxolitinib exposure become even hyperinflammatorylike in patients (26). Remarkably, the diminishing effects ofJAK inhibition on NK cell function in vitro are almostcompletely reversible, as the cytotoxic potential of previouslydrug-exposed NK cells against K562 cells was restored to levelsseen in solvent-exposed NK cells (Fig. 4D). This also supportsour data that ruxolitinib doses up to 10 mmol/L do not inducecell death in NK cells (Supplementary Fig. S1).

Ruxolitinib impairs synapse formation and inhibits the JAK/STAT signaling pathway

Knowing that ruxolitinib affects NK cell cytotoxic activity invitro and in vivo, we next addressed whether impaired lyticsynapse formation potentially explains the reduced killing.Confocal microscopy helps to visualize the interaction betweenNK-92 effector and K562 target cells (Fig. 5A) and shows a clearreduction of effector-to-target cell interactions. To quantify theobserved reduction in lytic synapse formation, doublet forma-tion of differentially labelled NK-92 and K562 cells was

Figure 3.Ruxolitinib (ruxo) impairs NK cell maturation in MPN patients. For the analysis of the NK cell maturation status, gated NK cells were analyzed in the peripheral bloodtaken from healthy donors, age-matched healthy donors, MPN patients treated or not with ruxolitinib, as indicated. The receptors allowing definition of thematuration status of NK cells analyzed by means of flow cytometry were NKp46/NKG2D (A), CD11b/CD27 (B), and KIR/NKG2A (C). Data, individual frequencies ofthe respective cell populations (A–C, first four blots with the mean shown as line) as well as the mean � SEM of indicated populations (A–C, bottom graph;� , P < 0.05; �� , P < 0.01; �� , P < 0.001).






Figure 4.Ruxolitinib inhibits the functionalcapacity of primary human NK cells.A, freshly isolated human NK cellswere activated overnight with1000 U/mL IL2 (top) or NKp46(bottom) in the presence of theindicated concentrations of ruxolitinib(n ¼ 10; �� , P < 0.001). Degranulationwas detected 6 hours after coculturewith K562 in a 10:1 ratio. For analysis ofNK-mediated target cell killing, CFSE(0.5 mmol/L)-stained K562 and NKcells were cocultured for 6 hours in a10:1 (effector-to-target) ratio in theabsence or presence of increasingruxolitinib concentrations (n ¼ 17;�� , P < 0.001). After 6 hours ofstimulation with IL12 (50 ng/mL), IL18(100 ng/mL), and Brefeldin A(10 mg/mL), IFNg production of humanNK cells was analyzed by intracellularcytokine staining and subsequent flowcytometry (n¼ 9; �� , P < 0.001). B, toanalyze NK cell proliferation, isolatedprimary human NK cells were stainedwith 0.5 mmol/L CFSE and cultivatedfor 5 days with IL2 and the indicatedconcentrations of ruxolitinib.Histograms depict one representativeexperiment. Pooled data from sixindependent experiments are shownas mean � SEM in the graph on theright (n¼ 18; �� , P < 0.001). C, NK cellswere activated with IL2 and cultivatedin the absence or presence of 1 mmol/Lruxolitinib for 5 days. NK cell receptorexpression was analyzed on days0 and 5. Representative histogramsdepict a NK cell control staining on day0 (filled line), a ruxolitinib-untreatedsample on day 5 (solid line), and aruxolitinib-exposed sample on day5 (dashed line). Bar charts combinedata from NK cells of six donorstreated with 1 mmol/L ruxolitinib orvehicle control on day 0 (black bar)and on day 5 (gray bar; � , P < 0.05;�� , P < 0.001). D, NK cells were firstincubated overnight with (gray bar) orwithout ruxolitinib (black bar), whichwas subsequently washed out, andcells were then restimulated for 24hours with IL2. Killing anddegranulation assays were performedon both time points as mentionedbefore (n¼8; �� ,P<0.001). All resultsare shown as relative data comparedwith the vehicle control. ns,nonsignificant.


Sch€onberg et al.




analyzed by means of a flow cytometry–based assay (represen-tative staining shown in Fig. 5B). Indeed, ruxolitinib dosedependently impaired stable physical interaction between NKand K562 cells (Fig. 5C).

Finally, we investigated potentially affected phosphorylationlevels of proteins integrating cytokine signals. We primarilyfocused on downstream signaling components of the JAK path-way using phosphoflow technology. Compared with the IgGcontrol, the signals of pS6, pSTAT5, and (less intense) pERK wereupregulated by cytokine stimulation in control samples exposedto solvent, whereas ruxolitinib induced a dose-dependentdecrease of IL2-induced S6 and STAT5 but not of ERK phosphor-ylation in primary human NK cells (Fig. 5D and E). Of note,phosphorylation levels of the primary target kinases JAK1 andJAK3, which both are relevant for NK cell function, were notaffected by ruxolitinib as proven byWestern blot (SupplementaryFig. S2).

DiscussionThis is the first study that provides a detailed analysis of the

influence of ruxolitinib on NK cell biology in MPN patients.We show that ruxolitinib impairs the NK cell compartmentboth in vitro and in MPN patients. Ruxolitinib drasticallyreduces NK cell numbers in ruxolitinib-treated patients whencompared with drug-na€�ve individuals. Importantly, we pro-vide evidence that patients with low NK cell numbers duringruxolitinib exposure also have a higher rate of clinicallyrelevant infections. Of note, the majority of infections wereof viral origin, which are known to be governed by a functionalNK cell compartment. Reduced NK cell numbers in ruxoliti-nib-exposed patients may, at least in part, be due to defectiveNK cell maturation, explaining the time-dependent decreaseof NK cell numbers during ruxolitinib intake. InsufficientNK cell renewal is reflected by an increased ratio of pheno-typically immature to mature NK cells. It is known that NK cellmaturation critically depends on appropriate cytokine signals(27, 28), especially IL2/IL15. NK cell precursors lackingIL15Rb also show an NK cell differentiation defect leading tosevere immunodeficiency (29). Thus, it is tempting to specu-late that ruxolitinib interferes with cytokine signals requiredfor terminal maturation of NK cells, explaining the shift to animmature NK cell phenotype. Moreover, IL15R integratessignals via its common g-chain in committed NK cell precur-sors and drives differentiation from immature to mature NKcells, as well as IL15 supports mature NK cell survival (30–32).Thus, it is almost impossible to define in detail at the patientlevel at which differentiation stage ruxolitinib interferes withNK cell differentiation, but probably most of them are affect-ed. Mutations in JAK3 (33) and/or STAT5B (34) also cause asevere combined immunodeficiency (SCID) syndrome inhumans, characterized by very low numbers or absence ofNK cells, further supporting the critical importance of anactive JAK/STAT pathway for proper NK cell differentiation.Whereas data suggest that JAK3 is central for IL2/15-induc-ed STAT5 phosphorylation (35), ruxolitinib has beendescribed to be a predominant inhibitor of JAK1 and JAK2(36). However, it has also recently been demonstrated thatruxolitinib also affects other tyrosine kinases, such as JAK3and Tyk2 (37), thus potentially explaining defective NK cellmaturation in ruxolitinib-treated MPN patients. On a signal-

ing basis, we could confirm this idea by showing that ruxo-litinib strongly prevented cytokine-induced STAT5 phosphor-ylation, which is a well-known downstream target of JAKactivation. Of note, the IL2-induced phosphorylation ofJAK1 and JAK3 is not significantly affected, which is supportedby previous data showing that ruxolitinib may even (despitepotent inhibition of the kinase activity) induce hyperphop-shorylation of JAK2 (38).

Thus, we suggest that ruxolitinib inhibits mainly the JAK3/STAT5 pathway downstream of the IL2/IL15 receptors, whichis critical for both NK activation and differentiation. Wecannot completely rule out that JAK2-V617F-mutatedNK cells (39), which have been shown to derive from theJAK2-V617F-mutated lymphohematopoietic progenitor, aremore sensitive to ruxolitinib-induced NK cell inhibition thantheir normal counterparts. However, data from a patient withruxolitinib treatment interruption showing rapid numericalNK cell recovery as well as the normal NK cell frequencies inMPN patients without ruxolitinib argue against an intrinsicdifferentiation defect. The potential reversibility of the rux-olitinib effects is also supported by our in vitro findings thatNK cell function is completely restored upon drug removal.Of note, previous reports demonstrated that MPN patientshave reduced NK cell numbers (40, 41). Our data are incontrast to these findings, as we clearly demonstrate that MPNpatients do per se not have reduced NK cells numbers untilthey receive ruxolitinib. However, in these particular articles,NK cells were characterized by CD16 positivity in the lym-phocyte gate, whereas we defined NK cells by CD56þCD3�

expression. In adjunct, we demonstrate that CD56bright NKcell numbers are increased in MPN patients, which may, atleast in part, be due to the hyperinflammatory syndrome thepatients have, as well as by a constitutive activation of JAK2in clonal NK cells leading to CD56 upregulation. This changeis reverted by ruxolitinib back to the level seen in controlpersons, presumably as a consequence of its anti-inflamma-tory activity.

In addition, a clear functional deficit of NK cells isolatedfrom MPN patients could be observed, even if they were nottreated with ruxolitinib. Especially, the killing activity of MPNNK cells is reduced compared with that of NK cells fromhealthy donors. It could not clearly be shown that the func-tional impairment is further aggravated when patients aretreated with ruxolitinib, as the killing and degranulationactivity was already at a very low level in MPN patients. Theobserved trend, however, in downmodulating the functionalactivity of NK cells is further supported by decreased expres-sion levels of various NK cell activation markers in ruxoliti-nib-exposed MPN patients. Ruxolitinib-induced NK cell dys-function is supported by the reduced frequency of NKp46-expressing NK cells when compared with age-matched con-trols and MPN patients without JAK inhibitor therapy,although NKp46 expression is regulated in an age-dependentmanner (42).

Our in vivo observations are supported by various in vitrodata. We provide clear evidence that ruxolitinib reduces thepotential of cytokine-mediated activation of NK cells fromhealthy donors, whereas the JAK-independent activationvia NKp46 remains unaffected. Inappropriate NK cell activa-tion is mirrored by a reduced expression of NK cell activ-ation markers, such as CD16, CD69, NKG2D, NKp46, and






Sch€onberg et al.





granzyme B. Killing of target cells critically depends on thegeneration of a functional lytic synapse (43). Interestingly,ruxolitinib reduced the ability of NK cells to interact withtarget cells and generate lytic synapses, which may, in addi-tion to the reduced expression levels of the NK cell activationmarkers, explain the reduced killing capacity of ruxolitinib-exposed NK cells. Our data may also be of importance whenconsidering a recent article showing that JAK inhibitorsincrease susceptibility of tumor cells to NK cell–mediatedkilling (44). However, in this article, the authors focusedonly at JAK inhibitory effects on the tumor cell side, whereasthe impact of JAK inhibition on the immune-cell side also hasto be taken into account, as systemic JAK inhibition maycounteract the sensitizing effects on the tumor cell level byimpairing NK cell function. Moreover, in the context ofallogeneic stem cell transplantation, ruxolitinib has recentlybeen suggested as a potential therapeutic option of GVHD(45). Our results might also be considered in this contextbecause NK cells are critical for the graft versus leukemia effectafter allogeneic stem cell transplantation (46).

In summary, our data provide first evidence that the JAKinhibitor ruxolitinib affects key characteristics of humanNK cells, such as cytokine-induced expansion and killingvia an impaired cytokine-mediated NK cell activation, lead-ing to a reduced effector-to-target cell interaction, which is aprerequisite for most NK cell functions. Accordingly, NKcells are highly efficient in the recognition and killing ofvirally infected cells (47), and, as a consequence, NK celldeficiency leads to a high susceptibility to various infections(48). Intriguingly, ruxolitinib therapy is also associatedwith severe infections, among which disseminated tubercu-losis (7, 8), reactivation of hepatitis B (9), progressivemultifocal leukencephalopathy (10), toxoplasmosis retinitis(11), and Epstein-Barr virus-associated aggressive lympho-ma (J. Richter, Lund University, Lund, Sweden; personalcommunication, June 2014) are the most alarming ones.Moreover, reactivation of herpes simplex and varicella zos-ter infections in ruxolitinib-treated patients is frequent,similar to patients with an inherited functional NK celldeficiency (49).

Our data may help to better understand the increased rateof severe infections in ruxolitinib-treated patients by show-ing potent NK cell–suppressive effects and complementrecent reports on ruxolitinib-induced immune dysfunction(13, 50), which supports the idea that ruxolitinib is a potentanti-inflammatory compound. The data should also be con-

sidered when testing ruxolitinib in solid tumors, as NKcells are an important component of cancer immune surveil-lance (14).

Disclosure of Potential Conflicts of InterestS. Koschmieder reports receiving commercial research support from,

is a consultant/advisory board member for, and has provided experttestimony for Novartis. T.H. Brummendorf reports receiving commercialresearch grant from and is a consultant/advisory board member forNovartis. No potential conflicts of interest were disclosed by the otherauthors.

Authors' ContributionsConception and design: K. Sch€onberg, P. Brossart, D. WolfDevelopment of methodology: K. Sch€onbergAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K. Sch€onberg, J. Rudolph, M. Vonnahme, S. Para-mpalli Yajnanarayana, M. Hejazi, M. Uhrberg, W. Verbeek, S. Koschmieder,T.H. Br€ummendorf, P. Brossart, A. Heine, D. WolfAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): K. Sch€onberg, J. Rudolph, I. Cornez, M. Uhrberg,S. Koschmieder, D. WolfWriting, review, and/or revision of the manuscript: K. Sch€onberg, J. Rudolph,M. Vonnahme, S. Parampalli Yajnanarayana, I. Cornez, M. Hejazi, A.R. Manser,M. Uhrberg, W. Verbeek, S. Koschmieder, T.H. Br€ummendorf, P. Brossart,A. Heine, D. WolfAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): K. Sch€onberg, J. Rudolph, I. Cornez,W. Verbeek, S. Koschmieder, T.H. Br€ummendorf, P. Brossart, A. Heine,D. WolfStudy supervision: D. Wolf

AcknowledgmentsThe authors thank all the patients and volunteer blood donors and

gratefully acknowledge the excellent technical help of Heidrun Struppeck.The authors thank Prof. Waldemar Kolanus and Dr. Thomas Quast (Molec-ular Immunology and Cell Biology, LIMES, Bonn) for providing microscopyequipment.

Grant SupportThis work was supported by BONFOR for K. Sch€onberg and the Deutsche

Forschungsgemeinschaft (DFG WO1877/1-1) for D. Wolf.The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received October 29, 2014; revised February 11, 2015; accepted March 4,2015; published OnlineFirst April 1, 2015.

Figure 5.Ruxolitinib prevents lytic synapse formation with NK target cells and inhibits cytokine signaling in NK cells. A, representative confocal microscopy stainingof NK-92 and K562 cells in the presence or absence of ruxolitinib is shown. Cells were stained for perforin (green) and F-actin (red). Target cells areshown in blue. Bar, 100 mm; stars, synapses. B, for quantitative examination of synapse formation, a flow cytometry–based synapse assay was used.Representative dot plots showing doublets between CFSE-labeled NK-92 and Far Red–labeled K562 cells are given for the indicated ruxolitinibconcentrations. C, pooled data from FACS analysis are presented as mean� SEM (n ¼ 4; �� , P < 0.01; �� , P < 0.001). D, isolated primary human NK cells werestimulated overnight with 1000 U/mL IL2 and at the indicated concentrations of ruxolitinib. Signaling events were analyzed by phosphoflowtechnology using various phospho-specific antibodies. Staining for pS6, pSTAT5, and pERK is shown. FACS plots depict data from one representativeexperiment. E, phosphorylation signal for each phospho-protein (expressed as arcsinh ratio of MFI) is related to the respective IgG k control of thevehicle-treated cells, as represented by each color-coded squares of the heatmap (as explained on the phosphorylation scale on the right of the heatmap).Each colored line in the graph represents pooled data (average arcsinh MFI � SEM) for each phospho-protein related to the respective vehicle-treated sample [pS6 (blue), pSTAT5 (red), pERK (green); n ¼ 4; � , P < 0.05; �� , P < 0.001].






References1. Tefferi A. Myelofibrosis with myeloid metaplasia. N Engl J Med 2000;

342:1255–65.2. Schindler C, Levy DE, Decker T. JAK-STAT signaling: from interferons to

cytokines. J Biol Chem 2007;282:20059–63.3. Percy MJ, McMullin MF. The V617F JAK2 mutation and the myeloprolif-

erative disorders. Hematol Oncol 2005;23:91–3.4. HarrisonC, Kiladjian JJ, Al-AliHK,GisslingerH,WaltzmanR, Stalbovskaya

V, et al. JAK inhibition with ruxolitinib versus best available therapy formyelofibrosis. N Engl J Med 2012;366:787–98.

5. Verstovsek S, Mesa RA, Gotlib J, Levy RS, Gupta V, DiPersio JF, et al.A double-blind, placebo-controlled trial of ruxolitinib for myelofibrosis.N Engl J Med 2012;366:799–807.

6. Santos FP, Verstovsek S. JAK2 inhibitors for myelofibrosis: why are theyeffective in patients with and without JAK2V617F mutation? Anti-cancerAgents Med Chem 2012;12:1098–109.

7. Colomba C, Rubino R, Siracusa L, Lalicata F, Trizzino M, Titone L, et al.Disseminated tuberculosis in a patient treated with a JAK2 selectiveinhibitor: a case report. BMC Res Notes 2012;5:552.

8. Hopman RK, Lawrence SJ, Oh ST. Disseminated tuberculosis associatedwith ruxolitinib. Leukemia 2014;28:1750–1.

9. Caocci G,Murgia F, Podda L, Solinas A, Atzeni S, LaNasaG. Reactivation ofhepatitis B virus infection following ruxolitinib treatment in a patient withmyelofibrosis. Leukemia 2014;28:225–7.

10. Wathes R, Moule S, Milojkovic D. Progressive multifocal leukoencephalo-pathy associated with ruxolitinib. N Engl J Med 2013;369:197–8.

11. Goldberg RA, Reichel E, Oshry LJ. Bilateral toxoplasmosis retinitis asso-ciated with ruxolitinib. N Engl J Med 2013;369:681–3.

12. Hurwitz H, Uppal N, Wagner SA, Bendell JC, Beck JT, Wade S, et al. Arandomized double-blind phase 2 study of ruxolitinib (RUX) or placebo(PBO) with capecitabine (CAPE) as second-line therapy in patients (pts)with metastatic pancreatic cancer (mPC). J Clin Oncol 2014;32:5s, 2014(suppl; abstr 4000).

13. Heine A, Held SA, Daecke SN, Wallner S, Yajnanarayana SP, Kurts C, et al.The JAK-inhibitor ruxolitinib impairs dendritic cell function in vitro and invivo. Blood 2013;122:1192–202.

14. Vivier E, Tomasello E, Baratin M, Walzer T, Ugolini S. Functions of naturalkiller cells. Nat Immunol 2008;9:503–10.

15. Lanier LL. Up on the tightrope: natural killer cell activation and inhibition.Nat Immunol 2008;9:495–502.

16. Ortmann RA, Cheng T, Visconti R, Frucht DM, O'Shea JJ. Janus kinasesand signal transducers and activators of transcription: their roles incytokine signaling, development and immunoregulation. Arthritis Res2000;2:16–32.

17. Schonberg K, Hejazi M, Uhrberg M. Protocol for the clonal analysis of NKcell effector functions by multi-parameter flow cytometry. Methods MolBiol 2012;903:381–92.

18. Schonberg K, Sribar M, Enczmann J, Fischer JC, Uhrberg M. Analyses ofHLA-C-specific KIR repertoires in donors with group A and B haplotypessuggest a ligand-instructed model of NK cell receptor acquisition. Blood2011;117:98–107.

19. Cornez I, JoelM, TaskenK, Langmoen IA,Glover JC, Berge T. EGF signallingand rapamycin-mediated mTOR inhibition in glioblastoma multiformeevaluated by phospho-specific flow cytometry. J Neurooncol 2013;112:49–57.

20. Hazeldine J, Lord JM. The impact of ageing on natural killer cell functionand potential consequences for health in older adults. Ageing Res Rev2013;12:1069–78.

21. Fu B,Wang F, SunR, Ling B, Tian Z,WeiH.CD11b andCD27 reflect distinctpopulation and functional specialization in human natural killer cells.Immunology 2011;133:350–9.

22. Chiossone L, Chaix J, Fuseri N, Roth C, Vivier E, Walzer T. Maturation ofmouse NK cells is a 4-stage developmental program. Blood 2009;113:5488–96.

23. Juelke K, Killig M, Thiel A, Dong J, Romagnani C. Education ofhyporesponsive NK cells by cytokines. Eur J Immunol 2009;39:2548–55.

24. Anfossi N, Andre P, Guia S, Falk CS, Roetynck S, Stewart CA, et al. HumanNK cell education by inhibitory receptors for MHC class I. Immunity2006;25:331–42.

25. Shilling AD, Nedza FM, EmmT, Diamond S,McKeever E, Punwani N, et al.Metabolism, excretion, and pharmacokinetics of [14C]INCB018424, aselective Janus tyrosine kinase 1/2 inhibitor, in humans. Drug MetabDispos 2010;38:2023–31.

26. Tefferi A, Pardanani A. Serious adverse events during ruxolitinib treatmentdiscontinuation in patients with myelofibrosis. Mayo Clin Proc 2011;86:1188–91.

27. Colucci F, Caligiuri MA, Di Santo JP. What does it take to make a naturalkiller? Nat Rev Immunol 2003;3:413–25.

28. Miller JS,McCullar V. Human natural killer cells with polyclonal lectin andimmunoglobulinlike receptors develop from single hematopoietic stemcells with preferential expression of NKG2A and KIR2DL2/L3/S2. Blood2001;98:705–13.

29. Gilmour KC, Fujii H, Cranston T, Davies EG, Kinnon C, Gaspar HB.Defective expression of the interleukin-2/interleukin-15 receptor betasubunit leads to a natural killer cell-deficient form of severe combinedimmunodeficiency. Blood 2001;98:877–9.

30. Cooper MA, Bush JE, Fehniger TA, VanDeusen JB, Waite RE, Liu Y, et al. Invivo evidence for a dependence on interleukin 15 for survival of naturalkiller cells. Blood 2002;100:3633–8.

31. Vosshenrich CA, Ranson T, Samson SI, Corcuff E, Colucci F, RosmarakiEE, et al. Roles for common cytokine receptor gamma-chain-depen-dent cytokines in the generation, differentiation, and maturation of NKcell precursors and peripheral NK cells in vivo. J Immunol 2005;174:1213–21.

32. Ranson T, Vosshenrich CA, Corcuff E, RichardO,MullerW, Di Santo JP. IL-15 is an essential mediator of peripheral NK-cell homeostasis. Blood2003;101:4887–93.

33. Roberts JL, Lengi A, Brown SM, Chen M, Zhou YJ, O'Shea JJ, et al. Januskinase 3 (JAK3) deficiency: clinical, immunologic, and molecular analysesof 10 patients and outcomes of stem cell transplantation. Blood2004;103:2009–18.

34. Bernasconi A, Marino R, Ribas A, Rossi J, Ciaccio M, Oleastro M, et al.Characterization of immunodeficiency in a patient with growth hormoneinsensitivity secondary to a novel STAT5b gene mutation. Pediatrics2006;118:e1584–92.

35. Lin JX, LeonardWJ. The role of Stat5a and Stat5b in signaling by IL-2 familycytokines. Oncogene 2000;19:2566–76.

36. Quintas-Cardama A, Vaddi K, Liu P, Manshouri T, Li J, Scherle PA, et al.Preclinical characterization of the selective JAK1/2 inhibitor INCB018424:therapeutic implications for the treatment of myeloproliferative neo-plasms. Blood 2010;115:3109–17.

37. Zhou T, Georgeon S, Moser R, Moore DJ, Caflisch A, Hantschel O.Specificity andmechanism-of-action of the JAK2 tyrosine kinase inhibitorsruxolitinib and SAR302503 (TG101348). Leukemia 2014;28:471–2.

38. Gorantla SP BK, Illert AL, von Bubnoff N, Peschel C, Duyster J. Ruxolitinibmediated paradox JAK2 hyperphosphorylation is due to the protection ofactivation loop phosphotyrosines from phosphatases. ASH 2013 2013;abstr 2847.

39. Bogani C, Guglielmelli P, Antonioli E, Pancrazzi A, Bosi A, Vannucchi AM.B-, T-, and NK-cell lineage involvement in JAK2V617F-positive patientswith idiopathic myelofibrosis. Haematologica 2007;92:258–9.

40. FroomP, Aghai E, Kinarty A, Lahat N. Decreased natural killer (NK) activityin patients with myeloproliferative disorders. Cancer 1989;64:1038–40.

41. Gersuk GM, Carmel R, Pattamakom S, Challita PM, Rabinowitz AP,Pattengale PK. Quantitative and functional studies of impaired naturalkiller (NK) cells in patients with myelofibrosis, essential thrombo-cythemia, and polycythemia vera. I. A potential role for platelet-derived growth factor in defective NK cytotoxicity. Nat Immun1993;12:136–51.

42. Almeida-Oliveira A, Smith-Carvalho M, Porto LC, Cardoso-Oliveira J,Ribeiro Ados S, Falcao RR, et al. Age-related changes in natural killer cellreceptors from childhood through old age. Hum Immunol 2011;72:319–29.

43. Orange JS. Formation and function of the lytic NK-cell immunologicalsynapse. Nat Rev Immunol 2008;8:713–25.

44. Bellucci R,NguyenHN,Martin A,Heinrichs S, Schinzel AC,HahnWC, et al.Tyrosine kinase pathways modulate tumor susceptibility to natural killercells. J Clin Invest 2012;122:2369–83.


Sch€onberg et al.




45. Spoerl S, Mathew NR, Bscheider M, Schmitt-Graeff A, Chen S, Mueller T,et al. Activity of therapeutic JAK 1/2 blockade in graft-versus-host disease.Blood 2014;123:3832–42.

46. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, TostiA, et al. Effectiveness of donor natural killer cell alloreactivityin mismatched hematopoietic transplants. Science 2002;295:2097–100.

47. Jost S, Altfeld M. Control of human viral infections by natural killer cells.Annu Rev Immunol 2013;31:163–94.

48. Orange JS. Natural killer cell deficiency. J Allergy Clin Immunol 2013;132:515–25; quiz 26.

49. Tong LX, Jackson J, Kerstetter J, Worswick SD. Reactivation of herpessimplex virus infection in a patient undergoing ruxolitinib treatment.J Am Acad Dermatol 2014;70:e59–60.

50. Parampalli Yajnanarayana S, St€ubig T, Cornez I, Alchalby H, Sch€onberg K,Rudolph J, et al. JAK1/2 inhibition impairs T cell function in vitro and inpatients with myeloproliferative neoplasms. Br J Haematol 2015 Mar 30.doi: 10.1111/bjh.13373. [Epub ahead of print].






2015;75:2187-2199. Published OnlineFirst April 1, 2015.Cancer Res Kathrin Schönberg, Janna Rudolph, Maria Vonnahme, et al. NeoplasmsJAK Inhibition Impairs NK Cell Function in Myeloproliferative

Updated version

10.1158/0008-5472.CAN-14-3198doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/03/31/0008-5472.CAN-14-3198.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/75/11/2187.full#ref-list-1

This article cites 47 articles, 15 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/75/11/2187.full#related-urls

This article has been cited by 15 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/75/11/2187To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-3198

http://cancerres.aacrjournals.org/content/suppl/2015/03/31/0008-5472.CAN-14-3198.DC1

http://cancerres.aacrjournals.org/content/75/11/2187.full#ref-list-1

http://cancerres.aacrjournals.org/content/75/11/2187.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/75/11/2187


jak inhibition impairs nk cell function in ...microenvironment and immunology jak inhibition impairs...

Documents