Background

Initiatives for the standardization of the Antinuclear Antibody assay on HEp-2 cells in Brazil began in 2000, with the First Brazilian Consensus for Antinuclear Antibody (ANA) Determination (BCA HEp-2) in Goiânia [1]. This initiative was successful in disseminate as standardized classification of ANA patterns in the country. In addition, BCA HEp-2 initiative inspired the launching of the International Consensus on ANA Patterns initiative in 2014. The present paper reports on the V BCA HEp-2 dedicated to the harmonization of the BCA HEp-2 with the ICAP guidelines. Aiming to keep coherence with the current international literature, the present report will use the recently consensed ICAP designation for the ANA test, i.e., indirect immunofluorescence assay on HEp-2 cells (HEp-2 IIFA), despite the fact that the designation recommended by the BCA HEp-2 is ANA-HEp-2 (see below).

The First BCA HEp-2 focused mainly on the definition of criteria for interpreting and reporting the test, elaboration of a classification algorithm for HEp-2 IIFA patterns based on topographic groups (nuclear, nucleolar, cytoplasmic and mitotic apparatus), and on the definition of general guidelines for the test procedure. At the occasion, it was established as conceptual framework that the pattern recognition and classification should be fundamentally based on morphological features [1].

The II BCA HEp-2, carried out in 2002 in Goiânia, defined the guidelines for issuing the descriptive report of the test. It was established that definite staining of any cell compartment, including the cytoplasmic, would be granted a positive test report. It was recognized that the coexistence of more than one autoantibody specificity in the sample can yield overlapping patterns and, therefore, the II BCA HEp-2 defined the group of mixed patterns. In addition, the II BCA HEp-2 defined preliminary clinical relevance and associations related to several HEp-2 IIFA patterns [2].

In 2007, the III BCA HEp-2 focused efforts on encouraging Brazilian laboratories to adopt quality control programs for the HEp-2 IIFA test, as well as educational programs. The relevance of conjugate titration was highlighted as a crucial step to adjust the reading system and harmonize the titer of results among different laboratories [3, 4].

In 2012, at the IV BCA HEp-2, in Vitória, three new patterns were included in the BCA HEp-2 classification algorithm: the cytoplasmic rods and rings pattern, the nuclear quasi-homogeneous pattern and the nuclear CENP-F pattern. Among the guidelines, the screening dilution was recommended at 1/80 and further serum dilution to at least 1/640. Recommendations were made regarding the pattern reproducibility in different commercial HEp-2 slide preparations and pertaining to the use of alternative methodologies for autoantibody screening, as well as for detection of specific autoantibodies [5].

The International Consensus on ANA Patterns (ICAP) was launched in São Paulo, in 2014, during the 12th International Workshop on Autoantibodies and Autoimmunity (IWAA). The purpose of the initiative is to discuss and promote consensus regarding patterns observed in HEp-2 IIFA, extending the concept and guidelines of the Brazilian ANA Consensus to specialists around the world [5], but always considering also the standardization initiatives from other countries [6,7,8,9].

The aim of ICAP was to establish global guidelines for the nomenclature and classification of HEp-2 IIFA patterns [10]. A group of 13 specialists from academic institutions in different countries were selected, with individual groups being responsible for studying the relevant literature and presenting a proposal to classify the HEp-2 IIFA patterns in each cell compartment. The session was attended by international specialists from academic, clinical laboratory and industry settings, who participated actively in the discussion. The underlying conceptual criteria were mainly based on the morphological details, but also considered in some cases the target autoantigens or target organelles. The ICAP classification algorithm mirrors that of the BCA HEp-2, but contains some peculiar aspects [10]. One innovation of ICAP is that each HEp-2 IIFA pattern was ascribed an alphanumeric code (Anti-Cell pattern) from AC-1 to AC-29. These codes allow for fast access and reference to all patterns classified under the ICAP algorithm. The ICAP recommendations have been published in successive papers [10, 11] and in the official website (www.anapatterns.org). This page has high penetration worldwide, currently presenting versions in ten languages, including Portuguese. The website www.anapatterns.org has been visited by more than 108,000 users in 171 countries, since its creation on May 2015. Recently, the ICAP team has published on detailed information on the clinical relevance of each of the 29 distinct HEp-2 IIFA patterns [12]. ICAP has broad international acceptance and its recommendations have been adopted around the world by academic institutions, clinical laboratories and the industry of kit manufacture. Currently, the insert of several HEp-2 cell slide kits contains the ICAP classification algorithm.

In view of the worldwide acceptance of ICAP, nowadays adopted in most countries, the specialists of the BCA HEp-2 group understood that it would be appropriate to seek harmonization between both initiatives. To this end, the V BCA HEp-2 was organized with the purpose of reviewing the recommendations and algorithm of the BCA HEp-2 from the perspective of the ICAP guidelines. This paper presents the recommendations of the V BCA HEp-2, enabling Brazilian clinical laboratories to update the recommendations as well as to provide an update for clinicians who order and interpret the test. This report is also available online at the Brazilian Consensus website (www.hep-2.com.br).

Methods

On August 27, 2016, during the XXXIII Brazilian Rheumatology Congress held in Brasília, 20 experts on the HEp-2 IIFA test participated in a full-day workshop dedicated to undertaking an analytical comparison between the BCA HEp-2 and the ICAP algorithms and recommendations. Participants included HEp-2 IIFA experts from academic and clinical laboratories across the country. Objectives of the meeting were: 1) to identify potentially diverging and similar points, devising strategies for harmonization between both initiatives; 2) to determine adequacy of the current and possible alternatives for a correct denomination of the ANA HEp-2 test; and 3) to ratify and update previous recommendations from BCA HEp-2. Each topic was introduced to participants, followed by extensive discussion until agreement, as in previous sessions of the BCA HEp-2. The standard strategy adopted in the former sessions of the Brazilian Consensus on ANA Patterns is based on an initial period of free and thorough discussion of each item on the agenda. In most cases, this discussion naturally leads to consensus on each item. In very few cases, in which no natural consensus is attained, a further discussion session is promoted. If any divergence persists, we generally postpone the decision to the next meeting of the group, recommending full discussion of the subject in the meantime. Very rarely, we had to promote a voting process for final decision. This procedure has been clarified in the previous publications of the group. In the present version of the meeting, a natural consensus was attained for all points in the agenda.

Recommendations

Harmonization of BCA HEp-2 with ICAP recommendations

The V BCA HEp-2 recommends the harmonization with the ICAP pattern classification algorithm available at www.anapattenrs.org and summarized on Fig. 1. Accordingly, all BCA-HEp-2 algorithms were adapted to match as much as possible the ICAP algorithm. In addition, the appropriate ICAP alphanumeric code (Anti-cell indications - AC) was added to each pattern, which contributes for consistency of the scientific information at an international level. ICAP has so far classified 30 patterns (Fig. 1) of autoantibodies on HEp-2 cells, which were ascribed the respective alpha-numeric AC codes, starting at AC-0, which corresponds to the negative pattern, ending in AC-29, which corresponds to the compound pattern of anti-topoisomerase I antibodies [10, 12, 13].

Fig. 1
figure 1

International Consensus on ANA Patterns classification algorithm (www.anapatterns.org)

Full size image

By means of minor adjustments, the ICAP algorithm was harmonized into the BCA HEp-2 original algorithm. The incorporation of AC codes corresponding to all HEp-2 IIFA patterns recognized by ICAP (Fig. 1) in the V BCA HEp-2 classification tree is depicted in Fig. 2. Based on the ICAP classification, the nucleolar patterns are allocated within the nuclear group. As the conceptual framework for the pattern classification in BCA HEp-2 is based on the analysis of fluorescence across four cell regions (nucleus, nucleolus, cytoplasm and mitotic apparatus), the V BCA HEp-2 chose to keep the nucleolar patterns as a distinct group, with the understanding that this arrangement does not contradict the ICAP conceptual framework and contributes for harmonization of both initiatives (Fig. 1).

Fig. 2
figure 2

V Brazilian Consensus for Autoantibodies on HEp-2 cells, classification algorithm

Full size image

The group of nuclear spekled patterns in the ICAP algorithm is not split into the subgroups with stained and non-stained mitotic chromatin, respectively. Instead, ICAP algorithm signals with an icon the patterns with stained plate. The V BCA HEp-2 maintains the original subdivision by understanding that this favors laboratory technicians in pattern analysis, facilitating the classification and improving the learning process. This division is also useful for the interpretation of possible autoantibodies associated with each pattern. As an example, the nuclear centromeric pattern in the ICAP classification tree is connected directly to the nuclear patterns box, and it is not classified within the nuclear speckled group. The V BCA HEp-2 maintains the cetromeric pattern (AC-3) linked to the nuclear speckled branch. In fact, the ICAP classification algorithm has several HEp-2 IIFA patterns directly linked to the root box (Fig. 1). However, the V BCA HEp-2 decided to maintain the classification guideline, in which individual HEp-2 patterns are linked to subgroups that derive from the root boxes (Fig. 2). Such tiered hierarchic algorithm contributes to the learning process and day-to-day operation by assisting in the comprehension and memorization of the several patterns.

The V BCA HEp-2 maintained the nuclear discrete dots pattern within the subgroup of speckled patterns with non-stained metaphase plate. This contrasts with the ICAP algorithm, which links this pattern directly to the root box of nuclear patterns. In addition, ICAP discriminates nuclear discrete dots patterns into two distinct patterns: multiple discrete nuclear dots (AC-6) and few discrete nuclear dots (AC-7) [10]. The AC-7 pattern has a low positive predictive value for systemic autoimmune diseases and is frequently reported in healthy individuals [14]. It is characterized by the decoration of 1 to 6 discrete nuclear dots resulting from the recognition of p80 coilin autoantigen [14] or SMN (survival of motor neuron) protein [15]. The AC-6 pattern (multiple discrete nuclear dots) is characterized by the recognition of 6 to 15 discrete nuclear dots of heterogeneous size and fluorescence in interphase cells [16, 17]. This pattern is associated with reactivity against the autoantigens Sp100 [15] or MJ/NXP-2 [18] and is observed in different autoimmune diseases such as Primary Biliary Cholangitis, dermatomyositis and other systemic rheumatic diseases [19].

The accumulated experience of the successive editions of the BCA HEp-2 has identified significant difficulty for analysts in the counting of the discrete dots and discrimination of few versus multiple discrete nuclear dots. This is especially true because the number and size of nuclear bodies stained by anti-p80 coilin and anti-sp 100 antibodies may vary, depending on the cellular substrate in use. Thus, although an experienced observer can safely suggest the most likely pattern in most cases, the number of points per nucleus does not seem to be an absolute criterion in distinguishing both patterns [3]. Therefore, since the third BCA HEp-2, the recommendation is that the rare discrete nuclear dots pattern (AC-7) and multiple discrete nuclear dots (AC-6) pattern are equally reported as “discrete nuclear pattern (AC6/AC7)”.

The pleomorphic speckled nuclear pattern (PCNA) (AC-13) remains in the group of nuclear speckled patterns with negative metaphase plate. In ICAP, this pattern is linked directly to the root nuclear box, and classified as expert level, which corresponds to the BCA HEp-2 optional report level. This is in contrast with the BCA HEp-2 recommendation that assigns the PCNA as a mandatory report pattern. This recommendation was maintained by the V BCA HEp-2, due to its significance in diagnosing systemic lupus erythematosus. An additional divergence refers to the CENP-F pattern, which is classified as a nuclear pattern in the ICAP algorithm and as a mixed pattern in the BCA HEp-2.

It is important to note that the above related divergent points between ICAP and BCA HEp-2 concern the hierarchical strategy for the pattern classification across the algorithms, but do not affect fundamental aspects, such as the definition and description of the patterns, as well as their associations with specific autoantibodies and clinical correlations. In that sense, laboratory analysts and clinicians will find a harmonical set of information in both algorithms, but at this point the BCA HEp-2 experts feel that the BCA HEp-2 algorithm provides further support in pattern identification and classification. In fact, the particularities of the BCA HEp-2 algorithm provide further support in pattern identification and classification.

Regarding cytoplasmic patterns, ICAP also recognizes the fibrillar and speckled subgroups, in the same way as BCA. However, mitochondrial and polar patterns are directly linked to the cytoplasmic patterns root box, whereas BCA HEp-2 classifies them within the secondary box of speckled cytoplasmic patterns (Fig. 2).

In the mitotic pattern group, ICAP includes the NuMA-like pattern (AC-26), characterized by nuclear fine speckled pattern plus spindle fibers decoration directly linked to the mitotic patterns root box. In fact, ICAP does not have now a classification group exclusively dedicated to mixed or compound patterns. In contrast, the BCA HEp-2 classifies the NuMA pattern (AC-26) as a compound pattern. The concept of compound patterns refers to patterns composed of staining of more than one cellular compartment and caused by a single autoantibody system targeting a single autoantigen, as is the case for NuMA-like pattern in which anti-NuMA antibodies react with the nucleoplasm and with the mitotic apparatus. ICAP and BCA HEp-2 fit the spindle fiber pattern (AC-25) within the mitotic patterns root box, side by side with other mitotic patterns such as the centrossome (AC-24) and the intercellular bridge (AC-27) patterns. However, BCA HEp-2 discriminates two closely related patterns, the NuMA 2 pattern (with extensive staining of the spindle fibers, anaphase intercellular bridge and the midbody) and the spindle fiber pattern (exclusive staining of the spindle fibers). The V BCA HEp-2 acknowledges and follows ICAP [10] by replacing the term “centriole pattern” with the term centrosome pattern (AC-24), considering the possible recognition not only of centrioles but also of other integrated components at this cellular domain [20].

Still in the mitotic apparatus group, ICAP adds a new pattern, the mitotic chromosomal envelope pattern, the only pattern classified by ICAP (AC-28) not yet recognized by BCA. This pattern is described as rare in discoid lupus erythematosus, chronic lymphocytic leukemia, Sjögren’s syndrome and polymyalgia rheumatica [21, 22]. The V BCA HEp-2 decided that participants should evaluate this novel pattern for further discussion in the future.

ICAP does not include some of the patterns classified by the BCA HEp-2, such as the reticulated coarse speckled nuclear pattern and the quasi-homogeneous speckled nuclear pattern. In addition, ICAP does not include the concept of compound patterns and, thus, does not include the compound patterns associated with anti-Ku (nuclear coarse speckled plus nucleolar homogeneus with a peripheral staining of the metaphase plate) and the P-ribosomal-associated pattern (dense fine speckled cytoplasmic plus weak homogeneous nucleolar patterns) [5, 10, 11]. The V BCA HEp-2 understood that these patterns should be kept in the Brazilian classification tree, since they have been useful for the interpretation of findings, for the establishment of possible associations and indicating further testing for specific autoantibodies. For a better inter-laboratory communication, in analogy with ICAP AC codes, the BCA created a transient alpha-numeric coding system (BAC: Brazilian anti-cell autoantibodies) to classify these patterns not yet recognized by ICAP. The main items that are peculiar to each of the two Consensuses are detailed in Table 1.

Table 1 Main differences between the Brazilian Consensus for Autoantibodies on HEp-2 cells (BCA) and the International Consensus on Ana Patterns (ICAP)
Full size table

The compound pattern comprising nuclear fine speckled with similar staining on the metaphase plate, nucleolar organizing region (NOR) and cytoplasmic fine speckled staining, was recently classified by the ICAP as AC-29, so called Topo I-like pattern, which is directly linked to the nuclear pattern root box [23]. This pattern, originally described by DELLAVANCE et al. [24] and not present in the former version of the BCA HEp-2 algorithm, is now incorporated in its upgraded algorithm, in the category of compound patterns. Table 2 shows six compound patterns recognized by the BCA HEp-2, three of which are already classified by ICAP, although not classified as compound patterns. Nevertheless, the V BCA HEp-2 ascribes the same AC code used in ICAP for these three patterns.

Table 2 Compound/mixed patterns of the Brazilian consensus for autoantibodies on HEp-2 cells
Full size table

As mentioned above, the V BCA HEp-2 confirms the concept of compound patterns as those in which a single autoantibody specificity yields distinct fluorescence patterns, be it in different cell compartments (nucleus, nucleolus, cytoplasm or mitotic apparatus) or different fluorescence patterns observed in the same compartment [24].

The V BCA HEp-2 sought to harmonize the Brazilian Consensus and the ICAP algorithms, maintaining the most valued peculiarities of the Brazilian classification algorithm and incorporating major advances from ICAP, with synergistic advantages for the interpretation of the patterns and autoantibodies associations.

Recommendations for naming the ANA test

Participants of the V BCA recommend the terminology “anti-cell antibody test”, but keeping the acronym ANA (in Portuguese, FAN, for fator antinúcleo). This trend had been predicted at the III BCA, when six options were recommended to designate the HEp-2 IIFA test, with the intention of gradually migrating to a single denomination. In the opinion of the BCA HEp-2 experts, ANA is the most used and most recognized term, as per rheumatologists and other specialists who commonly order this test. In addition, this is a recognized terminology in the official Clinical Pathology institutional agencies portfolio of tests in Brazil, which is of relevance for reimbursement purposes. In a recent study carried out by Silva and collaborators (2018), with participation of 53 Brazilian clinical laboratories, it was verified that most laboratories are using five of the six terms recommended by III BCA HEp-2. The term “ANA” was present in most of these denominations [25]. Recognizing the ICAP recommendation for naming the test as indirect immunofluorescence assay on HEp-2 cells (HEp-2 IIFA), used in the most recent publications [13, 23] and aiming to harmonize this report with the international current nomenclature, we will adopt in the term IIFA HEp-2 in this manuscript.

Choice of commercial brands and reproducibility of patterns

Recommendations of the IV BCA HEp-2 [5] regarding the distinct capacity of several commercial slides to express the various patterns was reiterated. The alert is emphasized due to a range of variation among different commercial substrates available in Brazil, which may affect the characterization and recognition of HEp-2 IIFA patterns. Variations can also be related to different lots of the same slide brand, something inherent to the manufacturing process.

Therefore, it is suggested that laboratories consider the possibility of using more than one brand for samples generating unexpected results, considering the limitation of some substrates in the expression of certain patterns [26]. Another recommendation is that, for each new lot or new brand, representative samples of different patterns should be tested, preferentially contemplating different cell compartments. The use of a control panel for validation of HEp-2 cell lots and brands is a measure that ensures greater reliability and safety of results.

It is the laboratory responsibility to address occasional problems with substrates that do not adequately express certain patterns, such as the fine speckled nuclear pattern associated with the presence of anti-SS-A/Ro antibodies (AC-4), the coarse speckled nuclear pattern associated with antibodies against RNP/Sm (AC-5), the pleomorphic nuclear pattern suggestive of antibodies anti-PCNA (AC-13), the compound pattern suggestive of the presence of anti-CENP-F antibodies (AC-14), the cytoplasmic fine speckled pattern suggestive of anti-Jo-1 (AC-20) antibodies, the cytoplasmic rods and rings pattern (AC-23), the anti-NuMA-1-like pattern (AC-26) and the anti-NuMA-2 pattern (AC-25) [10, 26].

Solid-phase methods for screening autoantibodies

BCA HEp-2 does not recommended the use of enzyme immunoassays and chemiluminescence assays in the autoantibody screening as a replacement for the HEp-2 IIFA method. This recommendation is due to the fact that the HEp-2 IIFA method allows for the detection of a large array of autoantibodies, including so far uncharacterized antibodies and some known autoantibodies not addressed in commercial multiplex reagents. Although the industry has promoted a virtuous progress in the improvement of several diagnostic methods, the use of solid phase-based methodologies has not yet surpassed the screening for autoantibodies by indirect immunofluorescence on HEp-2 cells in which experienced analysts are able to identify the different patterns indicating a large array of autoantibodies [5]. Careful morphological analysis of the fluorescence patterns directs the specific investigation of the most plausible autoantibodies present in the patient’s serum, conferring to the indirect immunofluorescence method on HEp-2 cells a unique and non-comparable dimension that currently surpasses other autoantibody testing modalities.

Screening dilution

An important issue with the HEp-2 IIFA method is the occurrence of positive results in an expressive proportion of healthy individuals (~ 13% in the Brazilian population at 1/80 dilution) [27]. In general, healthy subjects tend to have low titer (1/80 and 1/160), whereas most autoimmune patients show moderate (1/160 and 1/320) to high titers (≥ 1/640) [27, 28]. Thus, the V BCA HEp-2 recommends maintainance of the screening dilution at 1/80, as suggested in the previous Brazilian Consensus [5].

Titration of the conjugate

From a technical point of view, the determination of autoantibodies on HEp-2 cells depends on several factors. From those, microscope bulbs are of the utmost importance. According to a research across clinical laboratories in Brazil, 62% use 100-W bulbs, but the observed range encompasses 20, 30, 50, 120 and 200-W lamps [25]. This may represent an important factor of pattern recognition and end-titer disparity among laboratories. Other important aspects contributing to the heterogeneity of results are the protein/fluorescein ratio in conjugates, use of different control sera for system calibration (or its absence), lack of training and subjectivity inherent to the method. The conjugate titration is a step of great importance, allowing the adjustment of the system, ensuring recognition of the nominal titre of the control sera and greatly increasing the objectivity and accuracy of the test [3,4,5]. It is recommended that this procedure be performed for each new brand and lot of slide kit, thus, assuring consistency of titer in kits of different lots and even if a new slide brand is used. Low titer control serum samples represent an invaluable tool in this process [4]. The V BCA HEp-2 reinforces the relevance of this measure to improve the quality of the test, especially regarding the semiquantitative dimension that the test offers and adds value in clinical practice.

Assay quality control

The V BCA HEp-2 maintains the recommendation of need for rigorous quality control, already disclosed at the III and IV BCA HEp-2 [3,4,5]. Such a measure is fundamental in order to restrict the occurrence of false-positive reactions in individuals without autoimmune disease and as a measure to standardize results among different laboratories [25]. It is recommended that laboratories seek to participate in institutional quality programs and educational programs. Another parameter that integrates the quality control efforts is the accomplishment of the titration of the conjugate, as previously detailed.

The test report

The V BCA HEp-2 maintains the structured report recommended by the IV BCA HEp-2. There is one additional recommendation, that the ICAP code, whenever available, be added to the name of the pattern. A thoughtful description for each cellular compartment, which is a useful procedure for technical professionals performing and recording test results, favors appropriate pattern classification and clinical interpretation. Figure 3 shows an example of the HEp-2 IIFA report, classified by the V BCA and ICAP as AC-4 [10].

Fig. 3
figure 3

Representative laboratory report of the Indirect immunofluorescence assay on HEp-2 cells, expressing the AC-4 pattern. (a) Representative image. (b) Model report as recommended by the V BCA

Full size image

Conclusion

The recommendations of the V BCA HEp-2 are summarized in Table 3. The fifth edition represents another relevant initiative for the Brazilian process of standardization the HEp-2 IIFA test, initiated in the year 2000. Successful harmonization between recommendations by BCA HEp-2 and ICAP was achieved.. The remaining divergent points between both initiatives were clearly pointed out and perspectives for future harmonization were pinpointed. The Brazilian Consensus on ANA remains a key element allowing immunology laboratories to correctly perform and interpret the HEp-2 cell test in Brazil. In addition to harmonizing and incorporating several improvements from ICAP, the V BCA HEp-2 emphasized essential actions discussed in previous editions regarding quality control, strongly recommending that quality assessment actions must be implemented in the laboratory practice so that the HEp-2 IIFA test reaches the aimed high standard of quality and efficiency.

Table 3 Fifth Brazilian Consensus for Autoantibodies on HEp-2 cells: summary of recommendations
Full size table