Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2015 Mar 10;180(1):52–57. doi: 10.1111/cei.12556

How should a district general hospital immunology service screen for anti-nuclear antibodies? An ‘in-the-field’ audit

R Hira-Kazal *, P Shea-Simonds †, J L Peacock ‡, J Maher *,§,¶,✉
PMCID: PMC4367093  PMID: 25412573

Abstract

Anti-nuclear antibody (ANA) testing assists in the diagnosis of several immune-mediated disorders. The gold standard method for detection of these antibodies is by indirect immunofluorescence testing on human epidermoid laryngeal carcinoma (HEp-2) cells. However, many laboratories test for these antibodies using solid-phase assays such as enzyme-linked immunosorbent assay (ELISA), which allows for higher throughput testing at reduced cost. In this study, we have audited the performance of a previously established ELISA assay to screen for ANA, making comparison with the gold standard HEp-2 immunofluorescence test. A prospective and unselected sample of 89 consecutive ANA test requests by consultant rheumatologists were evaluated in parallel over a period of 10 months using both tests. ELISA and HEp-2 screening assays yielded 40 (45%) and 72 (81%) positive test results, respectively, demonstrating lack of concordance between test methods. Using standard and clinical samples, it was demonstrated that the ELISA method did not detect several ANA with nucleolar, homogeneous and speckled immunofluorescence patterns. None of these ELISANEG HEp-2POS ANA were reactive with a panel of six extractable nuclear antigens or with double-stranded DNA. Nonetheless, 13 of these samples (15%) originated from patients with recognized ANA-associated disease (n = 7) or Raynaud's phenomenon (n = 6). We conclude that ELISA screening may fail to detect clinically relevant ANA that lack defined specificity for antigen.

Keywords: anti-nuclear antibody, autoantibody, autoimmune testing, autoimmunity, audit

Introduction

Anti-nuclear antibodies (ANA) are extremely useful for the diagnosis of systemic lupus erythematosus (SLE) and systemic sclerosis (SSc), and are somewhat useful for diagnosis of Sjögren's syndrome and inflammatory myositis (IM; polymyositis and dermatomyositis) 1–3. These autoantibodies are also found in autoimmune hepatitis type 1 4, and are associated with increased risk of uveitis in patients with juvenile idiopathic arthritis (JIA) 5. Detection of ANA may precede the diagnosis of overt autoimmune disease, sometimes by several years 6. However, ANA may be detected in up to 30% of healthy individuals 7. Such false positive results are a major limitation to diagnostic utility 8 and result commonly from reactivity to dense fine speckled (DFS) 70 antigen 9.

In a recent position statement, the American College of Rheumatology has indicated that immunofluorescence-based testing remains the gold standard approach to ANA screening 10,11. Furthermore, they state that alternative diagnostic methods should demonstrate comparable performance to this standard. The status of immunofluorescence-based ANA testing as a reference method has been re-emphasized in a recent international consensus recommendation paper 12. Most commonly, such testing is conducted using fixed human epidermoid laryngeal carcinoma (HEp-2) cells, allowing the detection of up to 150 nuclear and cytoplasmic autoantibody types. However, HEp-2-based immunofluorescence ANA screening is slow, subjective and requires a high degree of interpretative skill. Moreover, HEp-2 testing is not amenable to standardization and may lack sensitivity for anti-Ro and ribosomal P autoantibodies 2. Owing to these limitations, combined with increasing workloads and budgetary restrictions, many laboratories have moved to ANA screening using solid-phase assay techniques, notably enzyme-linked immunosorbent assay (ELISA) and fluorescence-based enzyme immunoassays 13. Nonetheless, concerns remain about the level of clinical validation and surveillance that is required in respect of such high-throughput alternative methods 14, both prior to and after their introduction into clinical practice.

At Barnet hospital, we receive approximately 15 000 ANA requests annually. To maintain satisfactory turnaround, testing had been conducted using an ELISA system. However, colleagues in the Department of Rheumatology expressed their concerns about the ability of this assay to detect a sufficient proportion of clinically relevant ANA. Consequently, we audited the analytical performance of this assay ‘in the field’, making comparison with the gold standard HEp-2-based diagnostic test. While the relative sensitivity of both methods agreed with previous reports, we found that only HEp-2 testing detected a number of clinically relevant ANA that lacked defined target antigen specificity. This failure of ELISA testing to meet the audit standard resulted in the reinstatement of HEp-2 screening into our service.

Materials and methods

Study design

Eighty-nine unselected requests for ANA testing from rheumatology consultants were analysed in parallel using an ANA ELISA [ORG 600 ANA Detect (Orgentec Diagnostika GmbH, Mainz, Germany), comprising 26 human recombinant and highly purified native antigens, including all ANA of recognized clinical significance] and a HEp-2 indirect immunofluorescence assay (Nova Lite; Inova Diagnostics, San Diego, CA, USA). Samples were collected prospectively over a 10-month period from January 2013. Testing was also undertaken of ANA standards [IS2072 (homogeneous)/IS2100 (nucleolar)/IS2134 (centromere)], which were a kind gift of the Center for Disease Control (CDC) Biological Reference Reagents (Atlanta, GA, USA). Diagnosis was subsequently assigned from review of the electronic patient record and clinic letters.

Assay performance

Samples were analysed by UK state-registered biomedical scientists who were unaware of clinical details and results of other testing. Tests were performed according to the manufacturer's instructions, except as specified below. The cut-off for the ELISA test had been determined earlier in-house as an optical density (OD) ratio of ≥0·8 with respect to the kit control. This locally established threshold provides greater assay sensitivity than the manufacturer's recommended cut-off, which is ≥1·2 (borderline significance assigned to indices of 1·0–1·2). For HEp-2 testing, sera were screened at a dilution of 1/40 and, if positive, samples were diluted to obtain end-point titre (excluding nucleolar/centromere subtypes).

Samples that yielded discordant results in the screening ANA tests were tested for antibodies directed against extractable nuclear antigens [ENA (EliA Symphony, Freiburg, Germany), which detects antibodies to Sm, Ro52/60, La, U1 RNP, Scl-70, CENP-B and Jo-1] and double-stranded DNA (EliA dsDNA).

Statistical analysis

Specificity, sensitivity, positive predictive value (PPV) and negative predictive value (NPV) were determined with 95% confidence intervals (CI) using Confidence Interval Analysis software (version 2·2.0, University of Southampton, UK). Kendall's tau b rank correlation was calculated using Stata software (StataCorp, College Station, TX, USA).

Ethical issues

This was a routine audit of a laboratory method. Following enquiry to our local research and development (R&D) office and evaluation using Health Research Authority online decision tools (http://www.hra-decisiontools.org.uk/ethics/notresearch.html, accessed 12 September 2014), we have ascertained that the data presented do not constitute research, and that retrospective approval by a Research Ethics Committee is consequently not required.

Results

Comparison of results of ANA detection using ELISA and HEp-2 assay

Forty of 89 (45%) ANA tests were positive by ELISA, while 72 of 89 (81%) were positive by HEp-2 analysis (Table 1). The correlation between the antibody titre, as determined on HEp-2 cells and magnitude of reactivity upon ELISA testing, was statistically significant, but weak (Fig. 1; Kendall's tau b = 0·33, P < 0·001). Samples that tested positive in both assays had a PPV for ANA-associated disease of 0·54 (CI = 0·38–0·69), while samples that were negative in both assays had an NPV of 0·87 (CI = 0·74–0·93). Specificity, sensitivity, PPV and NPV of both ANA tests are indicated in Table 2.

Table 1.

Results of comparative anti-nuclear antibody (ANA) testing using human epidermoid laryngeal carcinoma (HEp-2)-based indirect immunofluorescence and enzyme-linked immunosorbent assays (ELISA)

HEp-2+ ELISA+ HEp-2+ ELISA− HEp-2− ELISA+ HEp-2− ELISA−
Number 37 (42%) 35 (39%) 3 (3%) 14 (16%)
Diagnosis
SLE 12 (13%) 4 (4%) 0 (0%) 0 (0%)
Sjögren's syndrome 6 (7%) 0 (0%) 0 (0%) 0 (0%)
Systemic sclerosis 2 (2%) 1 (1%) 0 (0%) 0 (0%)
Inflammatory myositis 0 (0%) 2 (2%) 0 (0%) 0 (0%)
Raynaud's phenomenon 0 (0%) 6 (7%) 0 (0%) 0 (0%)
Other diagnosis 17 (19%) 22 (25%) 3 (3%) 14 (16%)
 Rheumatoid arthritis 5 (6%) 7 (8%) 0 (0%) 2 (2%)
 Other autoimmune 8 (9%) 8 (9%) 2 (2%) 5 (6%)
 Non-autoimmune 4 (4%) 7 (8%) 1 (1%) 7 (8%)
Open in a new tab

SLE = systemic lupus erythematosus.

Figure 1.

Figure 1

Open in a new tab

Relationship between anti-nuclear antibody (ANA) titre as determined by human epidermoid laryngeal carcinoma (HEp-2) testing and enzyme-linked immunosorbent assay (ELISA) result. ELISA results are expressed as optical density ratio, corrected against a control. Analysis was performed on all 86 samples that were titrated to end-point dilution in the HEp-2 assay (two nucleolar pattern and once centromere pattern ANA were not titrated in this manner). Final diagnoses were systemic lupus erythematosus (SLE) (n = 15), Sjögren's syndrome (n = 6), systemic sclerosis (n = 2), inflammatory myositis (n = 2), Raynaud's phenomenon (n = 5) or other condition (n = 56).

Table 2.

Statistical indicators of anti-nuclear antibody (ANA) test performance

ELISA HEp-2
Diagnosis Parameter Mean CI Mean CI
SLE Specificity 0·62 0·51–0·90 0·23 0·15–0·34
Sensitivity 0·75 0·47–0·92 1·0 0·81–1·00
PPV 0·30 0·18–0·45 0·22 0·14–0·33
NPV 0·92 0·81–0·97 1·0 0·82–1·00
ANA-associated connective tissue disease Specificity 0·68 0·55–0·87 0·27 0·18–0·40
Sensitivity 0·74 0·53–0·88 1·0 0·88–1·00
PPV 0·50 0·35–0·65 0·38 0·27–0·49
NPV 0·86 0·73–0·93 1·0 0·82–1·00
Open in a new tab

ELISA = enzyme-linked immunosorbent assay; HEp-2 = human epidermoid laryngeal carcinoma; SLE = systemic lupus erythematosus; CI confidence interval; PPV = positive predictive value; NPV = negative predictive value.

Analysis of samples with discordant ANA test results

Of greatest clinical significance, we found that 38 (43%) samples yielded discordant results in ELISA and HEp-2 assays. Three samples were ELISAPOS HEp2NEG, two of which were from patients with an autoimmune disease (one weakly positive for anti-dsDNA antibodies, one positive for Ro antibodies). However, none of these three patients had a final diagnosis for which ANA screening is deemed useful, as defined above. However, 35 sera were HEp2POS ELISANEG, representing a spectrum of end-point titres and HEp-2 immunofluorescence patterns (Table 3). Thirteen of these samples were from patients with Raynaud's phenomenon (n = 6) or ANA-associated disease (n = 7), comprising SLE (n = 4), scleroderma (n = 1) or IM (n = 2). When tested for antibodies directed against ENA (n = 35) or dsDNA (n = 20), all these discordant samples proved negative.

Table 3.

Immunofluorescence pattern and titre of discordant human epidermoid laryngeal carcinoma (HEp2)+ enzyme-linked immunosorbent assay (ELISA)− serum samples

Pattern Titre Number No. with ANA-associated disease Number with Raynaud's phenomenon
Speckled 1/40 and 1/80 17 3 4
Speckled 1/160 and above 11 2
Nucleolar n.a. (screened at 1/40) 2 1
Speckled and homogeneous mixed >1/1280 1 1
Homogeneous 1/160 and above 3 1
Centromere n.a. (screened at 1/40) 1 1
Open in a new tab

n.a. = not available.

Testing of CDC standard sera using ELISA and HEp-2 assay

To compare assay performance further, standard sera were obtained from the CDC and were tested using both ELISA and HEp-2 based methods (Table 4). While the homogeneous and centromere pattern ANA were detected using both assays, the nucleolar pattern standard was only detected using the HEp-2 test.

Table 4.

Analysis of Center for Disease Control (CDC)-derived standard anti-nuclear antibody (ANA) sera

Reference control ELISA (ratio)* HEp-2 (titre)
Homogeneous ANA 5·2 Homogeneous titre >1/1280
Nucleolar ANA 0·3 Nucleolar pattern observed at 1/40 dilution (not titrated)
Anti-centromere ANA 2·0 Centromere pattern observed at 1/40 dilution (not titrated)
Open in a new tab
*

Positive samples are indicated by a ratio of ≥0·8 with respect to the kit control. ELISA = enzyme-linked immunosorbent assay; HEp-2 = human epidermoid laryngeal carcinoma.

Discussion

The ‘gold standard’ nature of the HEp-2 test has been emphasized in recent guidelines, in particular because of evidence that ELISAs may be less sensitive as a screening tool for ANA 10,12. Most laboratories continue to undertake HEp-2 testing, which allows detection of diverse ANA patterns in addition to other clinically significant autoantibodies (e.g. mitochondrial, multiple nuclear dots) 2. However, consensus on best practice is far from established, and a sizeable proportion of commercial and hospital laboratories screen for ANA using solid-phase assays. In support of the latter approach, some ELISAs have proved to be of equivalent clinical utility 15 or even more sensitive than HEp-2 testing for detection of ANA 16, while the poorly standardized nature of the HEp-2 test remains problematic.

Here, we describe the results of an audit in which we analysed the performance of our ELISA-based ANA screening test. Consultant staff in our rheumatology department were involved in all stages of the audit process. The audit standard was the result obtained from the same sample when tested using the HEp-2 immunofluorescence-based assay. The study was undertaken in a prospective manner in order to maximize accuracy of data collection and to reflect current rather than historical practice 17. In keeping with the majority of other studies, we found that ELISA was a less sensitive but more specific screening tool for SLE and other ANA-associated connective tissue diseases 11. However, testing by ELISA failed to detect a variety of ANA representing diverse immunofluorescence patterns, including the CDC standard anti-nucleolar antibody. All these ANA that were not detected by ELISA testing proved non-reactive against dsDNA or common ENAs, but were not tested further for other antigenic reactivity.

Against this background, the key question is the clinical significance of ANA that are detected by HEp-2 testing but not by ELISA. Some groups have argued that such antibodies are clinically irrelevant, as patients are indistinguishable from those who test negative in both assays 15. In agreement with this, it is stated on the College of American Pathologists website that it is ‘very uncommon’ to have a clinically significant positive on HEp-2 after a negative result in an ELISA-based assay 18. However, our experience does not tally with these conclusions, as we found that a sizeable proportion of these patients had an ANA-associated disorder, while a similar proportion had Raynaud's phenomenon. In the latter scenario, ANA positivity confers increased risk of progression to ANA-associated disease 7,19. Of related concern, a significant issue with false negative ANA testing using a variety of solid-phase assays has been reported in patients with SSc 20 and SLE 21–24. Furthermore, a recent study found a significant rate of false negative ANA ELISA results in children with JIA, although this did not associate with an increased risk of uveitis 25.

The clinical significance of ANA that are non-reactive with dsDNA or commonly used ENA panels also warrants comment. Examples of such antigens include ribosomal P, histones and the Ku antigen 26,27. It is well recognized that such ANA may be diagnostically useful in patients with drug-induced lupus (particularly with reactivity against histones) and autoimmune hepatitis. The audit data presented here indicate that ANA without defined antigenic target may also be clinically significant in rheumatological practice. Our audit demonstrated that ELISA screening for ANA failed to meet the standard of performance achieved using the HEp-2 assay. Consequently, we have implemented a change in practice such that ANA are now screened for using the HEp-2 assay alone, despite the requirement for additional staff time and skills. To minimize the impact of this change on test turnaround time, we have introduced an automated ANA reading system (Nova View, Inova Diagnostics) to enable automated high-throughput HEp-2 testing 28. While the findings of our audit cannot be generalized, our experience emphasizes the need to evaluate the diagnostic utility of solid-phase ANA assays in-house, making comparison with HEp-2-based testing.

Acknowledgments

We thank all rheumatology consultant staff at Barnet Hospital for useful comments and Carmina Sugui and Chris Taggart for assistance with sample testing and data collection. We acknowledge support from the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy's and St Thomas' NHS Foundation Trust and King's College London. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

Disclosure

There are no competing interests.

References

  1. Solomon DH, Kavanaugh AJ, Schur PH. Evidence-based guidelines for the use of immunologic tests: antinuclear antibody testing. Arthritis Rheum. 2002;47:434–444. doi: 10.1002/art.10561. [DOI] [PubMed] [Google Scholar]
  2. Mahler M, Meroni PL, Bossuyt X, Fritzler MJ. Current concepts and future directions for the assessment of autoantibodies to cellular antigens referred to as anti-nuclear antibodies. J Immunol Res. 2014;2014:315179. doi: 10.1155/2014/315179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Maher J. Role of the clinical immunology laboratory in disease monitoring. World J Immunol. 2013;3:18–30. [Google Scholar]
  4. Liberal R, Grant CR, Longhi MS, Mieli-Vergani G, Vergani D. Diagnostic criteria of autoimmune hepatitis. Autoimmun Rev. 2014;13:435–440. doi: 10.1016/j.autrev.2013.11.009. [DOI] [PubMed] [Google Scholar]
  5. Prieur AM, Chedeville G. Prognostic factors in juvenile idiopathic arthritis. Curr Rheumatol Rep. 2001;3:371–378. doi: 10.1007/s11926-996-0006-6. [DOI] [PubMed] [Google Scholar]
  6. Arbuckle MR, McClain MT, Rubertone MV, et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med. 2003;349:1526–1533. doi: 10.1056/NEJMoa021933. [DOI] [PubMed] [Google Scholar]
  7. Kavanaugh A, Tomar R, Reveille J, Solomon DH, Homburger HA. Guidelines for clinical use of the antinuclear antibody test and tests for specific autoantibodies to nuclear antigens. American College of Pathologists. Arch Pathol Lab Med. 2000;124:71–81. doi: 10.5858/2000-124-0071-GFCUOT. [DOI] [PubMed] [Google Scholar]
  8. Abeles AM, Abeles M. The clinical utility of a positive antinuclear antibody test result. Am J Med. 2013;126:342–348. doi: 10.1016/j.amjmed.2012.09.014. [DOI] [PubMed] [Google Scholar]
  9. Watanabe A, Kodera M, Sugiura K, et al. Anti-DFS70 antibodies in 597 healthy hospital workers. Arthritis Rheum. 2004;50:892–900. doi: 10.1002/art.20096. [DOI] [PubMed] [Google Scholar]
  10. American College of Rheumatology Position Statement. Methodology of testing for antinuclear antibodies. Available at: http://www.rheumatology.org/practice/clinical/position/ana_position_stmt.pdf (accessed 1 November 2014)
  11. Meroni PL, Schur PH. ANA screening: an old test with new recommendations. Ann Rheum Dis. 2010;69:1420–1422. doi: 10.1136/ard.2009.127100. [DOI] [PubMed] [Google Scholar]
  12. Agmon-Levin N, Damoiseaux J, Kallenberg C, et al. International recommendations for the assessment of autoantibodies to cellular antigens referred to as anti-nuclear antibodies. Ann Rheum Dis. 2014;73:17–23. doi: 10.1136/annrheumdis-2013-203863. [DOI] [PubMed] [Google Scholar]
  13. Nossent H, Rekvig OP. Antinuclear antibody screening in this new millennium: farewell to the microscope? Scand J Rheumatol. 2001;30:123–126. doi: 10.1080/030097401300162860. [DOI] [PubMed] [Google Scholar]
  14. Wiik AS, Gordon TP, Kavanaugh AF, et al. Cutting edge diagnostics in rheumatology: the role of patients, clinicians, and laboratory scientists in optimizing the use of autoimmune serology. Arthritis Rheum. 2004;51:291–298. doi: 10.1002/art.20229. [DOI] [PubMed] [Google Scholar]
  15. Copple SS, Sawitzke AD, Wilson AM, Tebo AE, Hill HR. Enzyme-linked immunosorbent assay screening then indirect immunofluorescence confirmation of antinuclear antibodies: a statistical analysis. Am J Clin Pathol. 2011;135:678–684. doi: 10.1309/AJCP6R8EELGODAYW. [DOI] [PubMed] [Google Scholar]
  16. Russell AS, Johnston C. Relative value of commercial kits for ANA testing. Clin Exp Rheumatol. 21:477–480. [PubMed] [Google Scholar]
  17. Maguire GA, Ginawi A, Lee J, et al. Clinical utility of ANA measured by ELISA compared with ANA measured by immunofluorescence. Rheumatology. 2003;48:1013–1014. doi: 10.1093/rheumatology/kep137. [DOI] [PubMed] [Google Scholar]
  18. Check W. 2009. Making sense of the ANA hodgepodge. CAP Today. September. Available at: http://www.cap.org/apps/portlets/contentViewer/show.do?printFriendly=true&contentReference=cap_today%2F0909%2F0909a_ANA_hodgepodge.html (accessed 1 November 2014)
  19. Moinzadeh P, Nihtyanova SI, Howell K, Ong VH, Denton CP. Impact of hallmark autoantibody reactivity on early diagnosis in scleroderma. Clin Rev Allergy Immunol. 2012;43:249–255. doi: 10.1007/s12016-012-8331-1. [DOI] [PubMed] [Google Scholar]
  20. Shanmugam VK, Swistowski DR, Saddic N, Wang H, Steen VD. Comparison of indirect immunofluorescence and multiplex antinuclear antibody screening in systemic sclerosis. Clin Rheumatol. 2011;30:1363–1368. doi: 10.1007/s10067-011-1766-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kroshinsky D, Stone JH, Bloch DB, Sepehr A. Case records of the Massachusetts General Hospital. Case 5-2009. A 47-year-old woman with a rash and numbness and pain in the legs. N Engl J Med. 2009;360:711–720. doi: 10.1056/NEJMcpc0807822. [DOI] [PubMed] [Google Scholar]
  22. Bonilla E, Francis L, Allam F, et al. Immunofluorescence microscopy is superior to fluorescent beads for detection of antinuclear antibody reactivity in systemic lupus erythematosus patients. Clin Immunol. 2007;124:18–21. doi: 10.1016/j.clim.2007.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Sjowall C, Sturm M, Dahle C, et al. Abnormal antinuclear antibody titers are less common than generally assumed in established cases of systemic lupus erythematosus. J Rheumatol. 2008;35:1994–2000. [PubMed] [Google Scholar]
  24. Emlen W, O'Neill L. Clinical significance of antinuclear antibodies: comparison of detection with immunofluorescence and enzyme-linked immunosorbent assays. Arthritis Rheum. 1997;40:1612–1618. doi: 10.1002/art.1780400910. [DOI] [PubMed] [Google Scholar]
  25. Sparchez M, Delean D, Samasca G, Miu N, Sparchez Z. Antinuclear antibody screening by ELISA and IF techniques: discrepant results in juvenile idiopathic arthritis but consistency in childhood systemic lupus erythematous. Clin Rheumatol. 2014;33:643–647. doi: 10.1007/s10067-014-2529-y. [DOI] [PubMed] [Google Scholar]
  26. Hoffman IE, Peene I, Meheus L, et al. Specific antinuclear antibodies are associated with clinical features in systemic lupus erythematosus. Ann Rheum Dis. 2004;63:1155–1158. doi: 10.1136/ard.2003.013417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Feki S, Frikha F, Ben Hadj Hmida Y, et al. Prevalence and diagnostic value of antinuclear antibodies without identified antigenic target: a retrospective study of 90 patients. Rev Med Interne. 2012;33:475–481. doi: 10.1016/j.revmed.2012.04.017. [DOI] [PubMed] [Google Scholar]
  28. Meroni PL, Bizzaro N, Cavazzana I, Borghi MO, Tincani A. Automated tests of ANA immunofluorescence as throughput autoantibody detection technology: strengths and limitations. BMC Med. 2014;12:38. doi: 10.1186/1741-7015-12-38. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES