SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

In this study, the immunological status of 61 patients with Fanconi anemia (FA) with advanced marrow failure before hematopoietic stem cell transplantation was analyzed by assessing the phenotype of peripheral blood lymphocytes, serum immunoglobulin (Ig) levels, and inflammatory cytokines. In patients with FA, total absolute lymphocytes (P < 0.0001), B cells (P < 0.0001), and NK cells (P = 0.003) were reduced when compared with normal controls. T cells (CD3), that is, cytotoxic T cells, naïve T cells, and regulatory T cells, showed a relative increase when compared with controls. Serum levels of IgG (P < 0.0001) and IgM (P = 0.004) were significantly lower, whereas IgA level was higher (P < 0.0001) than in normal controls. TGF-β (P = 0.007) and interleukin (IL)-6 (P = 0.0007) levels were increased in the serum of patients when compared with controls, whereas sCD40L level decreases (P < 0.0001). No differences were noted in the serum levels of IL-1β, IL-2, IL-4, IL-10, IL-13, IL-17, and IL-23 between FA subjects and controls. This comprehensive immunological study shows that patients with FA with advanced marrow failure have an altered immune status. This is in accordance with some characteristics of FA such as the proinflammatory and proapoptotic status. In addition, B lymphocyte failure may make tight and early immunological monitoring advisable. Am. J. Hematol. 88:472–476, 2013. © 2013 Wiley Periodicals, Inc.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

Fanconi anemia (FA) is an autosomal or X-linked recessive disease characterized by marrow failure, somatic malformations, and cancer proneness, primarily leading to acute myeloid leukemia (AML) and head and neck carcinomas [1].

The disease is due to lesions in one of the at least 15 genes currently known to be responsible for DNA repair mechanisms that render the cells sensitive to interstrand crosslinkers leading to a block in the G2 phase of the cell cycle. There is evidence that FA proteins, apart from their function in DNA repair, are also implicated in cytokine hypersensitivity, response to oxidative stress, and the immune response.

Scanty information is available on the immunological status in patients with FA. It has been suggested that some FA children have a generally increased infection susceptibility, not completely explained by neutropenia alone [2]. Castello et al. [3] found a grading of immunological defects in patients with FA and their family members. Recently, Myers et al. [4] demonstrated that patients with FA have reduced absolute numbers of NK and B cells and impaired cytotoxic function of NK and T cells. Patients with FA have an increased susceptibility to human papillomavirus (HPV)-associated cancer [5]. Holmgren et al. [6] showed a significant decrease of serological response in FANC-C mice after HPV vaccination, suggesting that in FA, a primary immune dysfunction may occur independently from bone marrow failure and may play a role in the pathogenesis of this disease.

Plasma levels of cytokines have also been investigated in patients with FA. TNF-α was thoroughly studied, and data show that FA cell lines produce high amounts of this cytokine [7] and that TNF-α is enhanced in plasma [8] as well as in the marrow of patients with FA in which it contributes to marrow failure [9]. In mouse models, TNF-α was shown to contribute to the progression of stem cells to AML [10]. Studies on IL-1β in patients with FA-A [11] showed overexpression of this cytokine in plasma of patients with a constitutively activated phosphoinositide 3-kinase-Akt pathway.

However, apart from these consistent data on TNF-α, findings on other cytokines were contradictory [7, 12, 13].

To address the issue of the immunological status of patients with FA with advanced marrow failure phase, we conducted a retrospective multicenter study on 61 patients with FA before hematopoietic stem cell transplant (HSCT) assessing the immunophenotype of peripheral blood lymphocytes, serum levels of immunoglobulins (Ig), and cytokines other than TNF-α that are involved in the immune response.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

Patients and controls

Frozen lymphocytes and sera of 61 patients with FA referred to Gaslini Children's Hospital, Genova, Italy, to the Department of Pediatrics, Leiden University Medical Center, Leiden, The Netherlands, and to the HSCT Unit of Hospital St. Louis, Paris, France, were used for the study. Recruited subjects included patients with FA who were assessed during the work-up process before HSCT and thus all were in advanced marrow failure stage. We were able to collect information about transfusion in 56/61 patients. Forty-two (75%) subjects were transfusion dependent for red cells and/or platelets. Median value of WBC in patients with FA was 3,075 × 109 L−1 (first quartile: 2.100 × 109 L−1; third quartile: 4.157 × 109 L−1). No patients younger than 4 years entered the study. No patients received specific treatments apart from transfusions at the time of sampling and all were infection free at least 2 weeks before sampling. Healthy controls were hospital controls, that is, children who were hospitalized for minor surgery or traumas and whose clinical indicators and laboratory markers turned out to be negative for infections and autoimmune and inflammatory diseases. Three different control groups were used: the first composed of 51 controls was for total lymphocyte and lymphocyte subset comparison, the second of 346 subjects for Ig, and the third of 23 individuals for cytokine serum levels measurement. Informed consent was obtained from patients and/or their relatives during the work-up process according to locally approved informed consent procedures.

Lymphocyte immunophenotyping

Lymphocyte subset analysis was performed by a six-color immunostaining panel and by a lysis and wash procedure. Briefly, 100 µL of EDTA-anticoagulated whole blood was incubated with monoclonal antibodies (mAbs) directed against surface-expressed antigens for 20 min at 4°C and lysed with FACS lysis solution (Becton Dickinson, NJ) for 10 min at room temperature. Data acquisition and analysis were performed on a FACSCanto flow cytometer (Becton Dickinson) equipped with two lasers (Argon 488 and HeNe 633), and six-color analysis was performed using FACS Diva™ software (Becton Dickinson). After applying fluorochrome-labeled mAbs (all produced by Becton Dickinson), the following peripheral blood lymphocyte subsets were investigated: CD3+ T cells, CD3+ CD4+ T-helper cells (Th), CD3+ CD8+ T-cytotoxic cells (Tc), CD16CD56+CD3 natural killer cells (NK), CD19+ B cells, CD4+ CD25high T-regulatory cells (Treg), CD3+ CD45RO+ T-memory cells, CD3+ CD45RA+ T-naïve cells, and HLA-DR+ T cells (T-activated cells). Leukocyte and lymphocyte absolute numbers were evaluated by a cell counter (Abbott Cell-Dyn Sapphire; Abbott Diagnostics, Abbott Park, IL).

Immunoglobulin levels

IgG, IgA, and IgM levels were determined during the clinical workup in each center by using standard methods.

Serum cytokine level measurement

Based on the knowledge of the proinflammatory status of patients with FA [9, 14], we determined the serum levels of a large panel of 10 proinflammatory cytokines involved in the immune response by adopting a bead-based immunoassay (FlowCytomix™ Comboplex Bender MedSystems) for the measurement of IL-1β, IL-2, IL-4, IL-6, IL-10, and IL-13 and of sCD40L by flow cytometry according to the manufacturer's instructions. Briefly, the protocol is based on a sandwich immunoassay combining two bead populations of different size, each one including multiple subset of beads differentiated by varying intensities of internal fluorescent dye. Acquisition was performed with FACSCanto cytometer. FlowCytomix Pro Software was used to calculate cytokine concentrations in each sample. Serum cytokine concentrations are expressed in pictograms per milliliter. IL-17, IL-23, and TGF-β1 were quantified by ELISA Kits (eBioscience, San Diego, CA).

Statistical analysis

Quantitative data describe medians and first and third quartiles. Comparisons between patients and controls were done with nonparametric Mann-Whitney U-test; the data were not normally distributed, and the homoscedasticity assumption was not fulfilled, and for these reasons, the nonparametric tests have been used. The normality of the distributions was evaluated by the Shapiro-Wilk test. Bonferroni's correction was applied to avoid multiple comparison error in the analysis of subgroups. The statistical package “Statistica” has been used for all the analyses (StatSoft Corp., Tulsa, OK).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

Sixty-one patients (31 males) were enrolled in this study. The median age was 8 years (range, 4–43 years). Complementation groups, available in 49 patients, were as follows: 39 FANC-A, three FANC-C, three FANC-D2, and one each for FANC-B, FANC-F, FANC-G, and FANC-L, respectively. Median absolute WBC counts were 3,075 × 109 L−1 (first quartile: 2,100 × 109 L−1; third quartile: 4157.5 × 109 L−1). All patients had platelets < 30 × 109 L−1, and some patients were already on red cell and/or platelet transfusions.

Absolute counts of total lymphocytes and lymphocyte subsets

In comparison with controls [absolute total lymphocyte count (ALC) median: 2.700 × 109 L−1; first quartile: 2.020 × 109 L−1; third quartile: 3.400 × 109 L−1], patients with FA had reduced ALC (median: 1.609 × 109 L−1; first quartile: 1.236 × 109 L−1; third quartile: 1.978 × 109 L−1; P < 0.000; Fig. 1A) and reduced proportions of B cells (CD19+, P < 0.0001; Fig. 1B) and NK cells (CD3CD16/56+, P = 0.003; Fig. 1C). Percentage of T cells (CD3+) was increased in patients with FA when compared with controls (P < 0.0001; Fig. 1D). As for the different T-cell subsets, in comparison with controls, patients with FA did not have a significantly different proportion of Th cells (CD3+CD4+, P = 0.90; Fig. 2A), but did have increased percentages of Tc cells (CD3+CD8+, P < 0.0001; Fig. 2B) and T naïve cells (CD3+CD45RA+, P < 0.0001; Fig. 2C). No difference was found in the percentages of the T memory compartment (CD3+CD45RO+, P < 0.25; Fig. 2D). The proportion of activated T cells (CD3+HLA-DR+) was decreased (P = 0.005; Fig. 2E), whereas that of T reg cells (CD4+CD25high) was increased (P < 0.0001; Fig. 2F) compared with controls (Table 1 and Supplementary Table I). In addition, when we divided the patients in different age subgroup (4–5, 6–11, 12–18, and >18 years), the results overlapped with those obtained in the whole population (data not shown). The comparison between A and non-A patients did not sort out any significant difference (data not shown).

image

Figure 1. Box plots of lymphocytes (A), B cells (B), NK cells (C), and T cells (D) in patients with FA and controls. Boxes represent median values with first and third quartiles. P values refer to the Mann-Whitney U-test.

Download figure to PowerPoint

Table 1. Comparison of Absolute Value of Lymphocyte Subsets Between Patients with FA and Controls
 Absolute counts (109 L−1)P-value
PatientsControls
Median (1st–3rd quartile)Median (1st–3rd quartile)
  1. P values refer to the Mann-Whitney U-test.

Lymphocytes1,601 (1236.0–1978.8)2700.0 (2020.0–3400.0)<0.0001
CD3+ CD4+ Th cells646.2 (471.9–822.8)1018.5 (729.8–1261.8)<0.0001
CD3+ CD8+ Tc cells608.1 (465.0–706.3)561.0 (421.9–727.4)0.72
CD3+ T cells1343.1 (1093.7–1663.4)1836.8 (1411.2–2273.8)<0.0001
CD19+ B cells107.9 (25.6–211.9)421.9 (314.9–603.2)<0.0001
CD16–56+ CD3 NK cells103.4 (56.0–174.2)357.2 (144.8–499.4)<0.0001
CD3+ HLA-DR+ T activated55.4 (21.1–103.1)139.1 (93.5–241.5)<0.0001
CD4+ CD45RO+ Th memory240.2 (176.3–265.8)313.6 (230.6–392.4)0.0006
CD4+ CD45RA+ Th naive418.1 (257.5–576.6)618.0 (490.8–860.9)0.0007
CD8+ CD45RO+ Tc memory80.0 (44.3–112.6)118.5 (84.4–170.7)0.0004
CD8+ CD45RA+ Tc naive459.0 (293.3–639.0)427.8 (311.1–621.0)0.9
CD3+ CD45RO+ T memory293.8 (196.7–367.4)536.5 (402.4–676.8)0.009
CD3+ CD45RA+ T naive892.0 (658.2–1184.8)1074.0 (831.1–1566.4)<0.0001
CD4+ CD25++ Treg33.2 (14.1–54.0)31.2 (19.2–39.8)0.92
image

Figure 2. Box plots of percentage distribution of T-helper (A), T-cytotoxic (B), T-naïve (C), T-memory (D), T-activated (E), and T-regulatory (F) cells in patients with FA and controls. Boxes represent median values with first and third quartiles. P values refer to the Mann-Whitney U-test.

Download figure to PowerPoint

Immunoglobulin serum levels

In patients with FA, IgG (P < 0.0001; Fig. 3A) and IgM (P = 0.004; Fig. 3B) levels were found to be significantly lower, whereas IgA level (P < 0.0001; Fig. 3C) was significantly higher than in the age-matched controls.

image

Figure 3. Box plots represent serum level of immunoglobulins G (A), M (B), and A (C) in patients with FA and controls. Boxes represent median values with first and third quartiles. P values refer to the Mann-Whitney U-test.

Download figure to PowerPoint

Cytokine serum levels

Serum level of TGF-β was increased in patients with FA (491 pg/mL) when compared with healthy controls (median: 453 pg/mL; 1st–3rd quartile: 465.7–556.0; P = 0.007; Table 2). The same observation was made for IL-6 (FA median: 8.5 pg/mL; 1st–3rd quartile: 0–52.3; controls median 0 pg/mL; P = 0.0007; Table 2). Soluble CD40 ligand was reduced in FA sera (median: 542 pg/mL) when compared with normal controls (5,102 pg/mL, P < 0.0001; Table 2). No differences were found between FA and control group with respect to all other tested cytokines (IL-1β, IL-2, IL-4, IL-10, IL-13, IL-17, and IL-23; data not shown).

Table 2. Cytokine Serum Level in Patients with FA and Controls
 PatientsControlsP-Value
 Median (1st−3rd quartile)Median (1st−3rd quartile) 
  1. P values refer to the Mann-Whitney U-test.

sCD40L (ng/ml)0.543 (0.00–2.774)5.102 (2.737-10.01)<0.0001
IL-6 (pg/ml)8.5 (0.00–52.3)0.00 (0.00-0.00)0.0007
TGF-β1 (pg/ml)491.3 (465.7–556.0)453.5 (407.1–497.4)0.007

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

The immunological profile of patients with FA has been previously evaluated in some descriptive studies [3, 4, 14] relying on a limited number of patients who were tested for few immunological parameters. Although our study is descriptive, it is based on a high number of patients (n = 61) who were assessed in a comprehensive way including an extensive lymphocyte subset testing, serum Ig level determination, and serum level measurements of a large panel of 10 cytokines. Overall, the immunological profile of patients with FA with advanced bone marrow failure is characterized by absolute lymphopenia with reduced B and NK cells, by a relative increment of T cells, that is, Tc, T naïve, and Treg cells, by a reduction of IgM and IgG and an increase of IgA serum levels, and by increased serum IL-6 and TGF-β and decreased sCD40L.

In a progressive disorder like FA, these findings cannot be considered as the true immunological profile of FA, but rather representing the immunological status detectable in an advanced marrow failure phase. However, our findings are consistent with others from the literature [3, 4, 14] and may impact on some clinical decisions concerning the management of patients with FA. The reduced B-cell counts and related serum Ig deficiency may prompt tight monitoring of Ig serum levels, and although FA subjects do not usually have an exceptionally high infection susceptibility, in subjects with low IgG levels and recurrent infections, Ig replacement can be considered. A more intense protection policy including vaccination of contacts and of patients with killed products and polysaccharide antigens can also be taken into account. As T-memory cells look less hampered than other subsets, it may be debated if vaccination with attenuated virus like measles, mumps, rubella, and varicella (MMRV) could be inserted in this protective plan.

This policy of infection prevention in FA may find some support by in vitro data indicating that stimulation of TLR4 and 8, mimicking the in vivo effect of infectious agents, induce overproduction of the myelosuppressive cytokine TNF-α [15] that may accelerate bone marrow failure.

Regarding the causes of the findings emerging from our study, we can only provide speculations. Absolute lymphopenia has no unequivocal explanation. Obviously, it may reflect, at least in part, the advanced marrow failure. Reasons why T cells look less affected than other subsets are not obvious. In fact, FA fetal CD34+ and iPS cells are known to exhibit the classical FA fragility features [16], which would point to an intrinsic deficiency of early FA hematopoietic cells. However, the fact that T cells early migrate out of the proapoptotic environment of the bone marrow might at least in part contribute to the reduced tendency of these cells to undergo apoptosis when compared with other subsets (B cells, committed hematopoietic cells) that are exposed for longer times to the harmful proapoptotic marrow environment.

Regarding naïve T cells, an additional explanation of their relative increase may derive from the homeostatic lymphopenia-induced proliferation [17], which occurs as compensatory mechanism when lymphocyte count drops and leads to cell cycle activation and is associated with the acquisition of a naive phenotype [18].

IgG and IgM serum levels were much lower in FA than in controls. This finding is consistent with the B lymphopenia and with a recent study [6], indicating that the FA mouse has an impaired development of the B-cell compartment as illustrated by suboptimal serological responses to standard vaccinations.

IgA levels were increased when compared with controls. This can be in accordance with prolonged stimulation of barrier defenses and with the increased levels of TGF-β1 and IL-6 as we observed in FA subjects, which stimulate the switch for IgA synthesis as part of the proinflammatory response [19].

As for the cytokine profile, as the role of TNF-α in marrow failure in FA is well established [9, 10, 20] and limited material was available, the analysis was restricted to the assessment of cytokines on which information was scarce.

The increase of serum TGF-β1 in patients with FA is in agreement with the proapoptotic state and with other findings of our study like the increase of regulatory T cells [21] and the low levels of IgG and IgM, the production of which is known to be inhibited by this cytokine [22]. It is worth noting that TGF-β1 has been shown to selectively inhibit the growth and differentiation of early HSCs [23, 24] and it cannot be excluded that this may contribute to marrow failure in FA.

The increased IL-6 levels are in agreement with the known proinflammatory status of FA that is also expressed by the increased production of IgA as we observed in our FA group.

Soluble CD40L is released mainly by platelets during their production, and thus, its low level may reflect marrow failure-related thrombocytopenia. It is worth noting that CD40L promotes activation and proliferation of B cells, Ig heavy chain switching, IgM response in vivo [25, 26], and the development of NK cells. Therefore, sCD40L reduction is consistent with the low B and NK cells and IgM levels as we found in our patients.

In summary, patients with FA with advanced marrow failure show immunological alterations that might reflect this condition and possibly the proapoptotic and proinflammatory status of FA. Although immunological studies initiated in an earlier phase may provide additional clinically relevant information, awareness of the alterations we found in advanced marrow failure may still be helpful to reduce the infectious risk of these patients.

Author Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

E.T. Korthof, R. Peffault de Latour contributed essential samples and analyzed the data; J. Svahn, G. Socié, J. Soulier, and V. Pistoia analyzed the data; A. Pistorio performed statistical analysis; F. Corsolini, A. Parodi, F. Battaglia, H. Moins-Teisserenc, M. Kok, R.G.M. Bredius, M. van Tol, and E.C.M. Jol-van der Zijde performed the research; P. Terranova performed the research and designed the research study; C. Dufour analyzed the data and wrote the manuscript; E. Cappelli designed the research study, analyzed the data, and wrote the manuscript.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

The authors thank ERG Spa, Rimorchiatori Riuniti, Cambiaso and Risso, SAAR Depositi Oleari Portuali, and Associazione Italiana Anemia di Fanconi (AIRFA) for supporting the activity of the Clinical and Experimental Haematology Unit, G. Gaslini Children's Hospital, Genova, Italy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Author Contributions
  8. Acknowledgments
  9. References
  10. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
ajh23435-sup-0001-supptab1.doc40KSupplementary Table I. Comparisons of percentage value of lymphocyte subsets between FA patients and controls. Tables represent median values with first and third quartiles in round parentheses. P values refer to the Mann-Whitney U test.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.