Response to rituximab in B-CLL patients is adversely impacted by frequency of IL-10 competent B cells and Fc[gamma]RIIIa polymorphism. A study of FCGCLL/WM and GOELAMS groups

OPEN

Citation: Blood Cancer Journal (2016) 6, e389; doi:http://dx.doi.org/10.1038/bcj.2015.115

Web End =10.1038/bcj.2015.115 http://www.nature.com/bcj

Web End =www.nature.com/bcj

LETTER TO THE EDITOR

Response to rituximab in B-CLL patients is adversely impacted by frequency of IL-10 competent B cells and FcRIIIa polymorphism. A study of FCGCLL/WM and

GOELAMS groups

Blood Cancer Journal (2016) 6, e389; doi:http://dx.doi.org/10.1038/bcj.2015.115

Web End =10.1038/bcj.2015.115 ; published online 22 January 2016

Rituximab (MabThera, Rituxan) in vivo mechanisms of action remain incompletely understood and could differ depending on the subtype of B-lymphoproliferative disorders. Rituximab has been shown to induce apoptosis, complement-mediated lysis, antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent phagocytosis (ADPC) in vitro and there is some evidence pointing towards an involvement of these mechanisms in vivo. Factors affecting rituximab response have been previously described including histology, tumor burden and rituximab pharmacokinetics.1 We previously observed that FcRIIIa-158 V/F polymorphism also impacts on clinical response. Because this polymorphism affects human IgG1 afnity for FcRIIIa expressed on both natural killer cells and macrophages, we postulated that ADCC was an important mechanism of rituximab activity in follicular lymphoma.2 New anti-CD20 antibodies, with higher afnity for FcRIIIa and ADCC such as obinutuzumab, have been therefore developed and are currently in clinical development.

Regulatory B cells have been identied in human and mice as a subset of B lymphocytes competent to secrete interleukin-10 (IL-10). These cells, also named B10 in mice, are characterized by their ability to modulate inammation, autoimmunity and adaptive and innate immune response through the production of IL-10.35 In mouse model, B10 cells would inhibit lymphoma cell clearance induced by anti-CD20 monoclonal antibodies (mAbs) through the regulation of monocyte Fc-mediated functions. Recently, it has been demonstrated that clonal chronic

lymphocytic leukemia (CLL) cells displayed such IL-10 competence and immunosuppressive functions.6 We therefore hypothesized that IL-10-competent B-CLL cells could inuence the clinical efcacy of rituximab in CLL patients.

A prospective, randomized phase II study (NCT01370772) including 140 patients was conducted between June 2012 and January 2013 in France. Treatment-naive patients (aged 1866 years) diagnosed with Binet stage C or active Binet stage A or B CLL were enrolled.7 Inclusion criteria are described in the supplementary Methods available on the Blood Cancer Journal website. In the experimental arm, standard udarabine-cyclophosphamide-rituximab (FCR) courses (six 28-day courses of rituximab: 375 mg/m2 D1 C1 and 500 mg/m2 D1 C2C6; udarabine: 40 mg/m2/d D24, cyclophosphamide: 250 mg/m2/d D24) were preceded by a prephase of rituximab: 500 mg on D0 and 2000 mg on D1, D8 and D15. Immuno-chemotherapy was planned to begin at D22. Primary end point results have been previously published.8

We considered that lymphocyte depletion after rituximab monotherapy assessed at D22 was a surrogate marker of in vivo rituximab activity, allowing thus to analyze inuence of IL-10-competent B-CLL cells on in vivo rituximab efcacy, in the 68 patients included in the experimental arm. Median lymphocyte count before the four doses of rituximab (D0) was 91.13 g/l (range:3.74497.40) and was 2.60 g/l (range: 0.14189.40) at the end of rituximab prephase (D22). Thus the median lymphocyte depletion after rituximab prephase (D22) was 95.1% (range: 77.0 to +99.9), among them 66% obtained more than 90% depletion. Patients characteristic and their distribution according to 90% lymphodepletion are presented in Table 1. No signicant correlation was found between 90% lymphodepletion and clinical (age, sex, Binet

Table 1. Parameters inuencing lymphocyte depletion induced by rituximab monotherapy in 68 CLL patientsLymphodepletion 490% (n = 44) Lymphodepletion 90% (n = 23) Univariate analysis Multivariate analysis

N (%) Median (IQR) N (%) Median (IQR) OR (95% IC) P-value AUC (95% CI) OR (95% CI) P-value

Age (years) 55.72 (51.3158.12) 53.99 (52.0757.41) 0.792

Men 31 (70.45) 18 (78.26) 0.68 (0.192.16) 0.693 Binet stage AB 36 (81.82) 16 (69.57) 0.51 (0.151.73) 0.404 ECOG 0 31 (70.45) 5 (21.74) 1.50 (0.416.29) 0.572 IGHV unmutated 25 (56.82) 16/22 (72.73) 0.50 (0.151.50) 0.324

Cytogenetic abnormalitiesDel(13q) 18/35 (51.43) 8/18 (44.44) 1.31 (0.414.29) 0.848 Del(11q) 7/42 (16.67) 6/23 (26.09) 0.57 (0.162.07) 0.560 Trisomy 12 2/33 (6.06) 2/14 (14.29) 0.40 (0.036.04) 0.572

2-microglobulin (mg/l) 39 (88.64) 2.90 (2.333.66) 22 (95.65) 2.76 (2.334.29) 0.857 CD38+ (%) 34 (77.27) 2.00 (0.0023.00) 17 (73,91) 10.50 (1.7526.00) 0.133

IL-10-competent cells (%) 32 (72.73) 2.30 (0.686.47) 15 (65.22) 9.51 (5.3515.70) 0.004 0.763 (0.6040.921) 0.83 (0.720.93) 0.002 FCGR3A 0.028 V/V 5 (11.91) 1 (4.55)

V/F 25 (59.52) 7 (31.82)

F/F 12 (28.57) 14 (63.63)

FCGR3A V carrier 30 (71.43) 8 (36.36) 4.23 (1.4313.42) 0.014 0.675 (0.5510.799) 4.95 (1.0727.48) 0.043

Abbreviations: AUC, area under the curve; ECOG, Eastern Cooperative Oncology Group Performance Status; IGHV, immunoglobin heavy-chain; OR, odds ratio; 95% CI, 95% of condence interval. Patients with less than 90% of lymphocyte count inhibition are used as the OR reference group.

Letter to the Editor

2

Figure 1. Effects of IL-10 B-CLL cells frequency and FcRIIIa-158V/F polymorphism on lymphocyte depletion induced by rituximab in CLL patients. (a) IL-10-competent B-CLL cells frequency correlated with IL-10 plasma level at D0 (P = 0.017). (b) Lymphocyte depletion observed after prephase of rituximab alone (D22) was statistically inuenced by IL-10-competent B-CLL cells frequency at D0 (P = 0.004). (c) Lymphocyte depletion between D0 and D22 was statistically higher for FcRIIIa-158V carriers patients (P = 0.014). (d) Receiver operating curve (ROC) generated using logistic regression for IL-10-competent B-CLL cells alone (area under the curve (AUC) = 0.763), FcRIIIa-158V/F polymorphism V carriers vs F/F (AUC = 0.675) and combination of IL-10-competent B-CLL cells and FcRIIIa-158V/F polymorphism V carriers vs F/F (AUC = 0.855).

stage, Eastern Cooperative Oncology Group Performance Status) and biological (IGHV mutation, cytogenetic abnormalities and 2-microglobulin) parameters. A subset of IL-10-competent B-CLL cells was detected in all patients tested (n = 47, median: 3.06% of CLL cells, range: 0.1229.55). High frequency of IL-10-competent cells among B-CLL cells was associated with a high IL-10 plasma level (Figure 1a, r = 0.40, P = 0.017) whereas the IL-10 plasma level did not correlate with the percentage of IL-10 negative B-CLL cells (data not shown). The frequency of IL-10-competent B-cells did not differ according to the characteristics of the patients, and was not signicantly different in CLL cases with unmutated vs mutated IGVH (median: 6.29%, range: 0.1215.83 vs median: 1.85%, range:0.2320.81, respectively). In addition, IL-10-competent B-cell frequency was not associated with cytogenetic alterations (del11q, del13q, trisomy 12). Such results were in agreement with a previous report.6 Univariate analysis showed that the frequency of IL-10-competent B-CLL cells adversely impacted on 90% lympho-depletion observed after rituximab prephase (D22) (Figure 1b, P = 0.004). In addition, IL-10-competent B-cell frequency was also

found to correlate with clinical response assessed 3 months after immuno-chemotherapy by FCR (complete response (CR) vs no-CR, P = 0.04). No correlation was found between lymphodepletion after rituximab prephase and response to immuno-chemotherapy (CR vs no-CR, P = 0.96). Thus, these results suggested that all CLL patients have a subpopulation of IL-10-competent B-CLL cells, which represent a variable percentage of B-CLL cells and exert a clinically signicant inhibitory effect on in vivo rituximab activity, probably through IL-10 secretion. Because FcRIIIa-158V/F polymorphism correlates with in vivo efcacy of rituximab in follicular lymphoma, we determined FcRIIIa-158V/F polymorphism in our cohort of patients. FcRIIIa-158V/F polymorphism was signicantly associated with 90% lymphodepletion (P = 0.028) and a normal lymphocyte count (o5g/l) (P = 0.028) at D22. This was also found for FcRIIIa-158V carriers (P = 0.014) (Figure 1c). FcRIIIa-158V/F polymorphism failed however to correlate with clinical response 3 months after immuno-chemotherapy by FCR. Thus, these results suggest that FcRIIIa-mediated immune functions play a critical role in clinical rituximab activity, but the impact of FcRIIIa-158V/F

Blood Cancer Journal

Letter to the Editor

3

polymorphism on immuno-chemotherapy response could be masked either by the high activity of immuno-chemotherapy or by direct inhibition of immune effector cells by chemotherapy. This is the rst report demonstrating the inuence of FcRIIIa-158V/F polymorphism in CLL patients. This is probably related to the fact that it is also the rst study analyzing such inuence in monotherapy context. These results could also explain clinical superiority of the rst Fc-glycoengineering anti-CD20 antibody exhibiting high afnity for FcRIIIa (obinutuzumab), compared with rituximab reported in CLL patients.9

Logistic regression analyses showed that only frequency of IL-10-competent B-CLL cells and FcRIIIa-158V/F polymorphism was associated with 90% lymphodepletion after rituximab prephase (odds ratio (OR) = 0.83; 95% condence interval (CI):0.720.93; P = 0.002 and OR = 4.95; 95% CI: 1.0727.48; P = 0.043, respectively). The receiver operating characteristic curve using IL-10-competent B-cells frequencies and FcRIIIa-158V/F polymorphism showed a highly discriminative power (area under the curve = 0.855; 95% CI: 0.7320.978), leading to predict patients who will have more than 90% of lymphodepletion after rituximab prephase (Figure 1d).

The experimental model suggested that IL-10-producing B cells would inhibit macrophage-mediated lymphoma depletion induced by anti-CD20 mAbs.10 Human macrophages express FcRIIIa which mediates ADCC induced by rituximab whereas FcRIIa, also expressed by macrophages, mediates ADPC. Data of this preliminary study indicate that IL-10-competent B-CLL cells frequency correlates with the ability of rituximab to induce B-CLL cells depletion. This effect seems especially important since IL-10-competent B-CLL cells frequency inuenced also clinical response after immuno-chemotherapy. The depletion of B-CLL cells was also independently inuenced by FcRIIIa-158V/F polymorphism with a better response for FcRIIIa-158V carriers, which have a better afnity for human IgG1. The inuence of FcRIIIa-158V/F polymorphism was however not found after immuno-chemotherapy, indicating that chemotherapy probably negatively impacts on FcRIIIa-expressing cells. In conclusion, all these data lead us to hypothesize that IL-10-competent B-CLL cells negatively regulate rituximab-mediated ADCC by macrophages in CLL patients, this effect being modulated by FcRIIIa-158V/F polymorphism. Strategies targeting IL-10-mediated inhibitory effects should be considered to improve rituximab efcacy.

CONFLICT OF INTEREST

Consultancy within the past 2 years: GC. Ownership interests (including stock options) in a start-up company, the stock of which is not publicly traded. Ownership interest (including stock options but excluding indirect investments through mutual funds and the like) in a publicly traded company. Research funding. Honoraria directly received from an entity: GC and CD. Paid expert testimony within the past 2 years. Any other potential nancial relationship (for example, holding a patent or receiving royalties). Membership on another entitys Board of Directors or its advisory committees (whether for prot or not for prot): CD.

ACKNOWLEDGEMENTS

We would like to thank Erika Nogu for her help supplied to statistical analysis. This work is supported by a public grant overseen by the French National Research Agency (ANR) as part of the Investissements dAvenir program (reference: ANR-10-LABX -53-01). This study was funded by GOELAMS group and Roche SAS (Neuilly, France).

A-L Gagez1, E Tuaillon2,3, R Cezar3, C Dartigeas4, B Mah5, R Letestu6, H Maisonneuve7, V Gouilleux-Gruart8, K Bollore2, E Ferrant9, T Aurran10, P Feugier11, S Leprtre12 and G Cartron1,13

1CNRS UMR 5235, Universit de Montpellier, Montpellier, France;

2INSERM U1058, Universit de Montpellier, Montpellier, France;

3Dpartement de Bactriologie-Virologie CHRU de Montpellier, Montpellier, France;

4Dpartement dHmatologie Clinique, CHRU de Tours, Tours, France;

5Dpartement dHmatologie Clinique, CHRU Nantes, Nantes, France;

6APHP, GHUPSSD, Hpital Avicenne, Service d'hmatologie biologique, Bobigny, France;

7Service de mdecine onco-hmatologie, CHD La Roche sur Yon, La Roche sur Yon, France;

8CNRS UMR 7292, Universit Franois Rabelais, CHRU de Tours, France;

9Dpartement dHmatologie Clinique, CHRU Dijon, Dijon, France;

10Centre Paoli Calmette, Marseille, France;

11Dpartement dHmatologie Clinique, CHRU Nancy, Nancy, France;

12Centre Henri Becquerel, Rouen, France and

13Dpartement dHmatologie Clinique, CHRU de Montpellier, Montpellier, France E-mail: mailto:[email protected]

Web End [email protected]

REFERENCES

1 Cartron G, Trappe RU, Solal-Celigny P, Hallek M. Interindividual variability of response to rituximab: from biological origins to individualized therapies.Clin Cancer Res 2011; 17: 1930.

2 Cartron G, Dacheux L, Salles G, Solal-Celigny P, Bardos P, Colombat P et al.

Therapeutic activity of humanized anti-CD20 monoclonal antibody and polymorphism in IgG Fc receptor FcgammaRIIIa gene. Blood 2002; 99: 754758.

3 Matsushita T, Yanaba K, Bouaziz JD, Fujimoto M, Tedder TF. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression.J Clin Invest 2008; 118: 34203430.

4 Yanaba K, Bouaziz JD, Haas KM, Poe JC, Fujimoto M, Tedder TF. A regulatory B cell subset with a unique CD1dhiCD5+ phenotype controls T cell-dependent inammatory responses. Immunity 2008; 28: 639650.

5 Iwata Y, Matsushita T, Horikawa M, Dilillo DJ, Yanaba K, Venturi GM et al. Characterization of a rare IL-10-competent B-cell subset in humans that parallels mouse regulatory B10 cells. Blood 2011; 117: 530541.

6 DiLillo DJ, Weinberg JB, Yoshizaki A, Horikawa M, Bryant JM, Iwata Y et al. Chronic lymphocytic leukemia and regulatory B cells share IL-10 competence and immunosuppressive function. Leukemia 2013; 27: 170182.

7 Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G, Dohner H et al.

Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008; 111: 54465456.

8 Lepretre S, Letestu R, Dartigeas C, Maisonneuve H, Aurran T, Feugier P et al.

Results of a Phase II Randomizing Intensied Rituximab Pre-Phase Followed By Standard FCR Vs Standard FCR in Previously Untreated Patients with Active B-Chronic Lymphocytic Leukemia (B-CLL). CLL2010FMP (for t medically patients): A Study of the French Cooperative Group on CLL and WM (FCGCLL/MW) and the Groupe Ouest-Est dEtudes Des Leucmies Aiges Et Autres Maladies Du sang (GOELAMS) [Poster]. 56th ASH Annual Meeting, San Francisco, CA, USA, 2014; 3329.

9 Goede V, Fischer K, Busch R, Engelke A, Eichhorst B, Wendtner CM et al.

Obinutuzumab plus chlorambucil in patients with CLL and coexisting conditions.N Engl J Med 2014; 370: 11011110.

10 Horikawa M, Minard-Colin V, Matsushita T, Tedder TF. Regulatory B cell production of IL-10 inhibits lymphoma depletion during CD20 immunotherapy in mice.J Clin Invest 2011; 121: 42684280.

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/ http://creativecommons.org/licenses/by/4.0/

Web End =by/4.0/

Supplementary Information accompanies this paper on Blood Cancer Journal website (http://www.nature.com/bcj)

AUTHOR CONTRIBUTIONSDesigned research: ET and GC; performed research: ALG, ET, RC, VGG, KB and

GC; collected data: CD, BM, RL, HM, EF, TA, PF, SL and GC; analyzed and interpreted data: ALG, ET, RC, VGG, KB and GC; performed statistical analysis: ALG and KB; wrote/reviewed the manuscript: ALG, ET and GC; approved manuscript: ALG, ET, RC, CD, BM, RL, HM, VGG, KB, EF, TA, PF, SL and GC.

Blood Cancer Journal