Complex patterns on HEp-2 indirect immunofluorescence assay in a large sample referred for anti-cell autoantibodies detection

Introduction The combination of patterns is a frequent and challenging situation in the daily laboratory routine of autoantibodies testing using HEp-2 cells indirect immunofluorescence assay (HEp-2-IFA). Recently, the Brazilian Consensus on Autoantibodies (BCA) named these combinations as complex patterns (CPs) and organized them into 3 subtypes: multiple, mixed, and composite. This study aimed to describe the most frequent combinations of HEp-2-IIF patterns according to this new nomenclature. Methods Routine HEp-2-IFA results reported in January and June 2017 were reviewed using the new BCA classification. Visual pattern recognition was performed by experts on HEp-2-IFA readings, using the International Consensus on Antinuclear Antibodies (ANA) Patterns (ICAP) and BCA recommendations. Results 54,990 serum samples from different patients were tested for ANA-HEp-2, and 11,478 (20.9%) were positive at a titer ≥ 1/80. Among these positive samples, 1,111 (9.7%) displayed CPs, divided into 95 different combinations. A higher proportion of CPs was observed in the pediatric age group. Multiple, mixed, and composite patterns were present in 85.3, 5.4, and 9.5% of the samples, respectively. In the multiple/mixed pattern group (n=1,005), double, triple, and quadruple combinations (ICAP/BCA codes) were observed in 97.7%, 2.2%, and 0.1%, respectively. The double nuclear pattern was the most prevalent combination observed (67.6%). The most common CPs registered were AC-4 (nuclear fine speckled) + AC-6,7 (nuclear discrete dots) (n=264); AC-2 (nuclear dense fine speckled) + AC-6,7 (n=201); AC-4+AC-8,9,10 (nucleolar) (n=129); and AC-3 (centromere)+AC-4 (n=124). All of these combinations were in the multiple subgroup. Conclusion Almost 10% of positive results in the HEp-2 procedure displayed CPs. Among the 3 subtypes of CPs proposed, the multiple pattern was the most prevalent, especially in the pediatric population. The AC-4, AC-2, and AC-6,7 were the most prevalent single patterns observed in the combinations described in this study. There was a significant association between age and the prevalence of most combined patterns. The AC-4+AC-6,7 combination was the most prevalent complex pattern detected regardless of the age group. The AC-2+AC-6,7 was more prevalent in younger individuals. The concepts involved in the CPs definition should add value to the reading and interpretation of the HEp-2-IIF assay.


Introduction
The Brazilian Consensus on Autoantibodies (BCA) recognizes 34 different positive staining patterns observed in the indirect immunofluorescence assay on HEp-2 cells (HEp-2-IFA).BCA-HEp-2 proposed the distribution of these patterns into 5 groups: nuclear, cytoplasmic, nucleolar, mitotic, and complex.The International Consensus on ANA Patterns (ICAP) recognizes 29 different positive patterns distributed into 3 groups: nuclear, cytoplasmic, and mitotic.Nucleolar staining is included in the nuclear group, and the complex patterns (CPs) are not depicted on the ICAP classification tree (1,2).
Anti-cell (AC) alpha-numeric codes were ascribed for each HEp-2 fluorescence pattern by the ICAP initiative.Therefore, the 29 ICAP-positive patterns received codes ranging from AC-1 to AC-29.All ICAP codes have been adopted by the BCA, except one, AC-28, described as mitotic chromosomal.On the other hand, there are 5 patterns recognized by the BCA that are not addressed by the ICAP (2,3).
The challenges in defining the fluorescence morphology from serum samples that present a combination of patterns have received the attention of the Brazilian consensus meeting participants since the II BCA in 2003.The terms "mixed pattern" or "composite pattern" were adopted to describe some situations where more than one cellular compartment is depicted due to the presence of a single antibody in the same sample or an overlap of patterns is observed due to the presence of more than one antibody (4).
This classification was updated at the VI BCA meeting that was held in September 2019 during the 36°Brazilian Congress of Rheumatology in Fortaleza-CE, Brazil.The group of CPs was introduced in the classification tree of BCA-HEp-2, thereby replacing the denomination of mixed patterns (1).
Based on the possible combinations of stained patterns and cell compartments distinguished by visual reading of HEp-2 cell slides, the CPs were organized into 3 subgroups: multiple, mixed, and composite.These subgroups were defined as follows: 1) multiple: a combination of patterns clearly identified on visual reading due to the presence of more than one autoantibody; 2) mixed: a mix of different patterns in the same cellular domain, not readily and identified at visual reading; and 3) composite: a combination of patterns due to the presence of a single autoantibody (1).
HEp-2 cells are considered an array for several target antigens.Serum samples that show reactivity against more than one antigen are not uncommon in daily laboratory routine.Under this circumstance, the correct interpretation of the HEp-2-IIF assay is challenging, requiring training and expertise (5)(6)(7).
Given the complexity and relevance of the CPs in daily laboratory practice, this study aimed to describe the most frequent pattern combinations observed on visual reading of HEp-2 cell slides in a large diagnostic medicine center in Brazil and to reclassify the mixed patterns recorded according to the new terminology proposed by VI BCA-HEp-2.

Study design
This is a retrospective and descriptive study in which the results of HEp-2-IFA carried out between January and June 2017 were reviewed.In this period, 57,265 consecutive serum samples from 54,990 patients were referred to the laboratory for routine autoantibody testing on HEp-2 cells.In cases where there was more than one serum sample from the same patient, only the first sample was included in the analysis.For each sample, gender and age were identified for further demographic analysis.It was not possible to obtain the clinical data of the participants.Patients and the public were not involved in the design, conduct, reporting, or dissemination plans of this research.

Report review
All reports with mixed/composite patterns in the period of the study were reviewed.This review included the fluorescence patterns released, the titers defined for each antibody observed, and their respective ICAP/BAC codes.As recommended by the BCA, our report structure includes the description of the staining morphology of each cellular compartment/structure: nucleus, nucleolus, cytoplasm, mitotic apparatus, and metaphase plate.Internal observations that were helpful in the interpretation of reading procedures were also included in the analysis worksheet and consulted whenever necessary.After performing these procedures, the mixed patterns were reclassified according to the new BCA-HEp-2 proposal (1).

HEp-2 indirect immunofluorescence testing
Anti-cell autoantibody tests were carried out on HEp-2 cell slides using the IIF technique according to the manufacturer's instructions (Euroimmun Medizinische Diagnostika, Lübeck, Germany).A confirmatory test was performed using another manufacturer (INOVA Diagnostics, San Diego, CA, USA) whenever necessary.Double serial dilution of positive samples was performed for endpoint-titer determination.Titers ≥ 1/80 were reported as positive for any fluorescence patterns observed in either the nucleus, cytoplasm, or mitotic apparatus as recommended by the BCA (3).Visual pattern recognition was performed by 3 experts on HEp-2-IIF readings.Pattern definitions were established according to ICAP and BCA-HEp-2 recommendations.

BCA and ICAP HEp-2-IIF pattern classification differences
On its decision tree, BCA-HEp-2 includes some patterns that are not recognized by the ICAP.These patterns received specific Brazilian Anti-Cell (BAC) codes, ranging from BAC-1 to BAC-5.The BAC-1, BAC-2, and BAC-5 are classified as composite CPs, while BAC-3 and BAC-4 represent elementary nuclear patterns.BAC-1 describes a nuclear fine speckled with nucleolar speckled pattern plus the decoration of the Nucleolar Organizing Regions (NOR) at the metaphase plate.BAC-2 describes a combination of nuclear coarse speckled, with nucleolar homogeneous and peripheral staining of the metaphase plate.BAC-5 presents a cytoplasmic dense fine speckled pattern with faint homogeneous staining in the nucleoli due to the presence of anti-P-ribosomal protein.BAC-3 and BAC-4 are defined as nuclear speckled Quasi-homogeneous and nuclear speckled reticular coarse patterns, respectively.
CENP-F-like (AC-14), NuMA-like (AC-26) and topoisomerase I-like (AC-29) are all patterns recognized by the ICAP that are included in the branch of the CPs on the Brazilian classification tree.Together with BAC-1, 2, and 5, they make up the composite subgroup of the CPs (1).BCA-HEp-2 does not recognize the ICAP AC-28 mitotic apparatus staining pattern.In the reports reviewed, some ICAP/BCA patterns were not detailed by the laboratory.The nucleolar homogeneous (AC-8), clumpy (AC-9), and speckled (AC-10) were reported in a grouped form as AC-8,9,10; the nuclear membrane continuous (AC-11) and dotted (AC-12) patterns were reported in a grouped form as AC-11,12; and the cytoplasmic fibrillar (AC-15), filamentous (AC-16) and segmental (AC-17) patterns were reported in a grouped form as AC-15,16,17.Likewise, the nuclear speckled with multiple (AC-6) and few discrete dots (AC-7) were reported as AC-6,7.Thus, although the BCA recognizes 34 positive patterns on HEp-2-IFA, due to the grouping of some patterns, only 27 were reported herein.

Autoantibodies detection data review
Patient's laboratory data reporting mixed patterns were reviewed to obtain information about specific autoantibodies testing in the samples assayed for HEp-2-IIF.Autoantibodies against DNA, Ro/SS-A, La/SS-B, Sm, U1-RNP, Jo-1, and Topoisomerase-I antigens are routinely tested in the laboratory if ordered by the clinician.In the period analyzed, anti-dsDNA antibodies were determined by IFI using Crithidia lucilae as substrate.Test positivity was considered for titles ≥1/10.Antibodies against the Extractable Nuclear Antigens (ENAs) encompassing Ro, La, Sm e RNP, Topoisomerase-I, and Jo-1 antigens were performed by fluorimetric enzyme-linked immunoassay (FEIA).Positivity and technical procedures for anti-ENAs testing were carried out according to the manufacturer's instructions (Thermo Fisher Scientific, Inc., MA, USA).

Statistical analysis
Categorical data were presented using frequency distributions and were analyzed using the chi-square test or Fisher's exact test when indicated.The Normality of the data was verified using the Kolmogorov-Smirnov test.For quantitative variables, the Mann-Whitney test was used for comparisons between two groups, and the Kruskal-Wallis test was used for comparisons involving multiple groups.In the latter case, post hoc pairwise comparisons were performed using the Bonferroni procedure.Analyses were performed using SPSS 22 (Statistical Package for the Social Sciences, IBM, USA) for Mac and Epi Info ™ for Windows version 7.2 CDC, USA.Statistical significance was indicated by a < 0.05.

Ethics statement
This study was approved by the Research Ethics Committee of the Centro Universitaŕio de Brası ́lia (UNICEUB) -Protocol number: 3.466.204/2019.According to the terms of this approval, all patient data and information were anonymized and deidentified before analysis and the results were presented in a grouped way so that individual results were not possible to be identified.Based on these, the absence of written informed consent was accepted by the Ethics Committee and no specific consent procedure was required for this study.

Results
During the 6 months studied herein, 54,990 different samples were tested for autoantibodies on HEp-2 cells, and 11,478 (20.9%) were positive at a titer ≥ 1/80.Of these, 1,111 (9.7%) displayed a combined staining pattern and were reclassified as CPs.The majority of the samples (n=10,367) were described using an ICAP/BCA single pattern.

Demographic data
Of the 54,990 sera, 74.3% were from female patients.Among all positive results (n=11,478), the proportion of samples from females was 85.1%.This amount was quite similar to the proportion of females in the complex-pattern group, which was 86%.
The age range of all participants was 1-102 years old and the median age was 43 years old.The age range of the 11,478 HEp-2-IFA-positive patients was 1-97 years.HEp-2-IFA-positive results were separated into a single-pattern group (n=10,367) and a complex-pattern group (n=1,111).The median ages of the singlepattern and complex-pattern groups were 42 and 38 years old, respectively.Participants in the complex-pattern group were younger than those in the single-pattern group (Mann-Whitney test, p <.001).

Combinations of patterns according to HEp-2 cells anatomy
The report structure used in our laboratory routine encompasses the description of the staining patterns in some cellular regions and structures: nucleus, nucleolus, cytoplasmic mitotic apparatus, and mitotic chromosome plate.
The patterns distribution of the 11,478 positive sera, according to cellular compartments was: nuclear (83.1%), cytoplasmic (4.0%), nucleolar (2.5%), and mitotic (0.7%).The combination of patterns or domains in the same sample (CPs) was observed in 9.7% of the patients.
Concerning the multiple and mixed subgroups, the nuclear region was involved in some of the reported combinations in 990 (98.5%) sera.Multiple/mixed patterns without nuclear staining were restricted to 15 out of all 1,005 samples.In these 1,005 samples, double (n=982), triple (n=22), and quadruple (n=1) combinations of patterns (ICAP/BCA codes) were observed in 97.7, 2.2, and 0.1%, respectively.

Single patterns and CPs according to ICAP and BCA proposals
Among the 11,478 positive results, nuclear fine speckled (AC-4), nuclear dense fine speckled (AC-2), nuclear quasi-homogeneous (BAC-3), CPs (as a group), and nuclear homogeneous (AC-1) were the 5 most prevalent single patterns, and they were observed in 37.7%, 21.3%, 10%, 9.7% and 6.8% of the samples, respectively.These 5 patterns accounted for 85.5% of all HEp-2-IFA positive results.Besides BAC-3, BAC-4 (nuclear reticular coarse) was another pattern recognized only by the BCA observed among the positive HEp-2-IFA results.Its frequency as a single pattern was 0.9% and in combination with another ICA/BCA pattern was 1,2%.We observed only one description compatible with BAC-5 (cytoplasmic dense fine speckled plus nucleolar homogeneous).The presence of BAC-1 (nuclear fine speckled with nucleolar speckled pattern plus the decoration of the Nucleolar Organizing Regions) was not possible to define because, in 2017, the nucleolar fluorescence was not discriminated by the laboratory.Finally, we did not find any matching description for BAC-2 (nuclear coarse speckled plus nucleolar homogeneous).We observed 95 different combinations of staining patterns in the 1,111 samples classified as CPs.From these, 81, 11, and 3 types of combinations were described in the multiple, mixed, and composite patterns, respectively.In the multiple/mixed subgroups (n=1,005), the AC-4 pattern was observed 644 times.The AC-6,7 pattern was present in 512 combinations.The AC-2 and AC-3 patterns were identified in 212 and 160 combinations, respectively.The vast majority of the AC-6,7 pattern was due to AC-7 (few nuclear dots); however, we did not separate these two patterns systematically, so it was not possible to determine accurately their prevalence rates.

Pattern combinations in the composite subgroup
The most common composite patterns (n=106) observed herein were AC-26 (NuMA-like) (n=99), AC-29 (Topo I-like) (n=6) and AC-14 (CENP-F) (n=1).In this review, it was possible to identify one serum with a description compatible with BAC-5 (AC-19 + AC-8,9,10), classified as multiple, since the nucleolar decoration was not specified.No report compatible with BAC-1 and BAC-2 was identified.In 2 samples, it was reported the combination of NuMA-like with cytoplasmic patterns (AC-18 and AC-21).Both samples were classified as multiple.In one sample with a history of positive anti-dsDNA, it was observed the metaphase plate was homogeneously stained.But as the nuclear staining was registered as fine speckled, it was classified as composite rather than mixed.

Single patterns and CPs prevalence according to age
The AC-4, AC-2, BAC-3, CPs, and AC-1 were the most prevalent patterns observed in 11,478 HEp-2-IFA-positive patients.For analyzing the prevalence of HEp-2-IIF elementary patterns and the CPs as a group according to age, the sample was divided into two groups: under 40-year-old (n=5,227) and over 40year-old group (n=6,251).The prevalence of female sex in each group was 83.3% and 86.6% respectively.
The AC-4 and AC-2 patterns and the CPs as a group were more prevalent in individuals under 40 years old (p <.0001).The prevalence ratios (RP) for these patterns in this age group were higher for CPs and AC-2 than AC-4.The BAC-3, AC-1, and the "other patterns" are more common in individuals ≥ 40 years old (p <.001).The RP varied from 0.56-0.77,indicating a lower risk for younger individuals to present these patterns.These data adjusted for sex are presented in Table 2.
Similar to what was done with single patterns, for the analysis of the association between the prevalence of CPs subtypes and age, the whole sample was divided into two groups: under 40 years old (n=582) and over 40 years old (n=529).The prevalence of female sex in each group was 82.3% and 89.9% respectively.
The results showed that AC-4+AC-6,7 and AC-2+AC-6,7 were more prevalent in the younger individuals (<40 years old).The chance of positivity for these combinations was almost double in this group of patients (prevalence ratio: 1,7-1,87).Combinations with the nucleolar pattern and the AC-26 pattern were more prevalent in individuals over 40 years old.The low prevalence ratios (0.56 and 0.48) observed indicated a lower chance of finding these combinations in people under 40 years old (p <.0001 for both CPs).The "other patterns" were more prevalent in individuals over 40 years old, but it was especially more prevalent in elderly people over 60 years of age, where it reached a proportion of 44.3%.No difference in prevalence was observed between the age groups for the AC-4+AC-3 combination.These data adjusted for sex are presented in Table 3.
For additional analysis of the association between the prevalence of some subtypes of CPs and age, the sample was divided into the same 4 age groups described previously: pediatric, young adults, adults, and elderly.
Considering the 3 subtypes of CPs proposed by the BCA, multiple patterns were observed in 177 of 184 sera (96.2%) of the pediatric group (0-19 years old, n=184).The proportion of mixed and composite patterns in this group was only 1.6% and 2.2% respectively.Taking the other groups together (n=927), the proportion of multiple, mixed, and composite was 83.2%, 5.8%, and 11%, respectively.The higher prevalence of multiple patterns in the pediatric group was significant (c 2 = 20.8,df=1, p <.001).
The prevalence of the AC-4+AC-6,7 combination in the pediatric group (44.0%, n=184) was higher when compared to the other groups (19.3%, n=927) (c 2 = 52.3,df=1, p <.0001).In elderly patients (≥ 60 years old, n=185), the "other patterns" accounted for almost half of the CPs (44.3%).The prevalence of these patterns in the elderly group was highly significant when compared to the prevalence in the group under 60 years old (n=926) (c2 = 35.9,df=1, p <.0001).The combination of nuclear fine speckled plus nucleolar was the second most prevalent (18.4%) in this group.By the way, the prevalence of this combination increased with age (p = .0007)(Table 3).Similarly, the proportion of "other patterns" gradually increased with age.

Antigenic specificities of the autoantibodies detected
From the 1,111 samples that depicted a combined immunofluorescence pattern, 242 (21.8%) were tested simultaneously for the presence of specific autoantibodies.Positive results for any of the requested antibodies were observed in 56 samples (23.1%).In only 20 (8.3%) samples more than one autoantibody was detected.
The 36 samples with only one autoantibody detected, presented a nuclear pattern associated with another immunofluorescence pattern due to an antigen not tested or unknown.The most frequent autoantibodies identified in this subgroup were anti-Ro (n=18), anti-dsDNA (n=6), anti-Topoisomerase-I (n=6), anti-Sm (n=4), and anti-RNP (n=2).The fluorescence pattern corresponding to the autoantibody detected was not identified in 10 samples (27.8%).The ICAP patterns related to Ro (AC-4) and RNP (AC-5) antigens were not identified in these samples.
In the 20 samples with more than one autoantibody detected, we observed the presence of 2 (n=14), 3 (n=5), and 4 autoantibodies (n=1).All of them are due to reactivities against dsDNA and ENA profile (Ro, La, Sm, U1-RNP).The fluorescence pattern expected was detected in 14 samples.In the 6 (30%) remaining samples, at least one expected pattern was not observed.All of these samples were classified as mixed patterns according to the new VI BCA proposal.

Discussion
National and international standardization initiatives have yielded valuable contributions to the reading and interpretation of HEp-2-IIF staining patterns, thereby paving the way for the standardization of reports and the improvement of the interpretation of the HEp-2-IFA results.The Brazilian HEp-2-IFA consensus movement started in 2001, and in its sixth edition, efforts were directed towards the organization of the mixed patterns, designated by the VI BCA-HEp-2 as CPs.CPs are described in the literature as mixed, multiple, composite, combined, overlapping, miscellaneous, or simultaneous patterns.To the best of our knowledge, this is the first article to specifically address this issue.
Based on the VI BCA recommendations, the reports of 54,990 HEp-2-IIF assays issued in the first half of 2017 were reviewed and all samples classified previously as mixed patterns were reclassified as CPs, considering the three subtypes proposed: multiple, mixed, and composite (1).Among the 11,478 HEp-2-IFA-positive results, we identified 1,111 (9.7%) samples with CPs, according to the VI BCA proposition.The CPs, as a group, were the fourth most common pattern observed.The AC-4, AC-2, and BAC-3 were the 3 most frequent single patterns identified in all positive HEp-2-IFA samples.In the literature, the prevalence of mixed patterns varies from 3.5 to 28.6% (5,(8)(9)(10)(11)(12)(13)(14).This wide range of positivity may be explained by differences in sampling, reasons for requesting the exam, as well as the degree of adherence to international recommendations on ANA testing and reporting.Prado et al., in a patient-based study with 269 systemic lupus erythematosus (SLE) patients, found that the multiple overlapping patterns were the third most frequent, present in 21.2% of the SLE patients.Only AC-1 (29.3%) and AC-4 (28.6%) single patterns were more prevalent than the combined HEp-2 patterns (14).In a hospital-based study involving 3,960 consecutive patients, following the AC-4, the mixed patterns were the second most prevalent pattern in 866 HEp-2-positive patients from several medical specialties (9).

Reports with multiple/mixed ANA patterns account for an i m p o r t a n t p a r t o f t h e d a i l y r o u t i n e w o r k u p o n autoantibody diagnostics.
Some combinations of patterns play a relevant role in the investigation and evaluation of autoimmune diseases (1,5,15).Wei et al., in a patient-based study with 4,583 subjects (3,510 with Systemic Autoimmune Rheumatic Diseases -SARD and 1,073 Health Individuals), found ANA positivity in 78.7% of SARD patients.AC-4, AC-1, AC-5, AC-8-9, and mixed patterns were the most frequent results in the SARD group.According to these authors, mixed patterns provided hints for systemic sclerosis and systemic lupus erythematosus (SLE) (12).Soldani and Hadalwar found 13.5% mixed patterns in 280 ANA-positive samples from 650 patients.Speckled (17.85%) and homogeneous (8.59%) nuclear patterns were the most frequent single patterns reported (10).The combinations of reticular and speckled cytoplasmic patterns with nuclear patterns such as AC-4, AC-1, and AC-3 were described in an expressive proportion of patients with autoimmune liver diseases (16).A distinctive combination of patterns, involving the concomitant presence of cytoplasmic and nuclear patterns, such as AC-21, AC-3, AC-6, and AC-12 can provide relevant diagnostic and prognostic information in patients with primary biliary cholangitis (15).
Regarding the HEp-2 cells anatomy and the topographic distribution of the patterns found in our study, the proportion of nuclear, nucleolar, cytoplasmic, mitotic, and mixed patterns reported was 83.1%, 2.5%, 4.0%, 0.7%, and 9.7%, respectively.These distributions of the HEp-2 patterns, regarding its cellular compartments, are in line with other publications, except for the cytoplasmic domain that appears overrepresented in some studies (9,11,17).According to some authors, the prevalence of cytoplasmic patterns increases with age (17).This trend was observed in this study and will be discussed later.The proportion of the single and mixed patterns according to HEp-2 cell domains in different studies is presented in Table 4.As in the ICAP proposal, the nucleolar patterns are arranged within the nuclear group, the truly nuclear combinations of patterns account for 82.2% of the CPs.The double nuclear (67.6%) and the nuclear plus nucleolar combination (14.6%) were by far the most common combinations observed (Table 1).Recognizing multiple antibody reactivities in the same cellular compartment can be a challenging task and requires training (6,7).

ICAP and BCA patterns classification differences
The BCA-HEp-2 incorporated the category "mixed patterns" into its classification tree in 2003.In its 6 th edition (1), BCA named this category as CPs and divided it into three subgroups: multiple, mixed, and composite.ICAP uses the terminology mixed or multiple to describe the situation in which more than one staining pattern is seen on the HEp-2-IIF assay in the same sample.The proposition of creating the composite patterns as a separate category did not reach consensus in the ICAP meetings (18,19).Unlike the BCA, the multiple/mixed patterns do not figure in the ICAP decision tree.Some other differences between these initiatives will be discussed below.
The Brazilian consensus distinguishes a higher number of HEp-2-IFA patterns than ICAP.This favors a higher number of pattern combinations.The BCA recognizes all the patterns proposed by the ICAP but one, the AC-28.On the other hand, there are five patterns recognized solely by the BCA, identified with specific Brazilian Anti-Cell codes (BAC) ranging from BAC-1 to BAC-5.The BAC-1, BAC-2, and BAC-5 codes describe a particular combination of patterns due to the presence of a single autoantibody and are classified as composite CPs.The BAC-1 pattern is a combination of nuclear and nucleolar staining patterns, associated with the presence of anti-RNA polymerase I/II antibodies.The BAC-2 code refers to the nuclear speckled and nucleolar homogeneous combination associated with the presence of anti-Ku antibodies.The BAC-5 is a combination of cytoplasmic dense fine speckled plus a nucleolar homogeneous, due to the presence of anti-ribosomal-P proteins.The BAC-3 (quasi-homogeneous) and BAC-4 (reticular coarse) are nuclear elementary patterns placed within the speckled group (1,3).
Besides these specific BCA patterns, in the last consensus meeting, the BCA proposed splitting the nuclear fine speckled pattern (AC-4) into AC-4a and AC-4b, for an optimal report (1).The AC-4a is described as a myriad of tiny discrete nuclear speckles observed in interphase cells with negative metaphase chromatin plate and is associated with the presence of Anti-SS-A/Ro60 antibodies (1,4).The AC-4b is described as a plain fine speckled   34).In our research, the most prevalent combinations seen in the pediatric group were the AC-4+AC-6,7 (44.0%) and AC-2+AC-6,7 (23.4%).Both prevalence rates were highly significant.We did not find similar report in the literature.
Considering the whole sample, the most common CPs identified were: AC-4+AC-6,7; AC-2+AC-6,7; AC-4+AC-8,9,10; AC-3+AC-4 and AC-26.As observed for the AC-4 and AC-2 single patterns, the combinations of these patterns with the discrete nuclear dots (AC-6,7) were more prevalent in the pediatric group and young adults.For the combined pattern AC-4+AC-8,9,10 and AC-26, a higher prevalence in the age group ≥ 40 years old was significant.The prevalence of different combinations increased with age.We observed that the proportion of "other patterns combinations" in elderly individuals reached 44.3%.As observed by Satoh et al., cytoplasmic patterns increased with age (17).In our study, the rate of positivity of cytoplasmic patterns increased progressively: 1.2%, 2.3%, 4.7%, and 7.4% in pediatric, young adult, adults, and elderly patients, respectively.ranges defined in this study.Cytoplasmic patterns increased with age.Therefore, the proportion of combinations involving the cytoplasmic domain was higher in individuals ≥ 60 years of age.
In the literature, the mixed patterns are cited without details about the patterns involved in the combinations.We identified reports of combinations containing mainly the patterns: nuclear speckled, nuclear homogeneous, nucleolar, and cytoplasmic (10,31,34) or citing combinations of stained cellular domains and the proportion of mixed patterns without specifying the patterns identified (11,12,17).We believe this is the first article that presents the mixed patterns specifying its compositions using an internationally standardized nomenclature.
The prevalence of the AC-6,7 and BAC-3 patterns in the CPs identified Among all 11,478 positive patterns, AC-4, AC-2, and BAC-3 were the most frequent elementary patterns observed.However, in the complex-pattern group, AC-4, AC-6,7, and AC-2 were the most common patterns identified in the several combinations described.The AC-6,7 was frequently seen in combinations with other patterns, mainly with AC-4 and AC-2 patterns.Considering that this pattern is not among the most prevalent elementary patterns in most frequent single pattern in the HEp-2-positive samples.We believe that this was due to a misinterpretation of the pattern when it was in combination with another nuclear pattern.The BAC-3 presents an intermediary morphological feature between the AC-1 and AC-2 patterns.Its reactivity reflects a heterogeneous profile of autoantibodies and unlike the AC-2 pattern, its presence is associated with several clinical conditions (39,40).Besides, the BAC-3 positive samples usually do not show high reactivity on HEp-2 cells and the presence of another antibody in higher concentration may make its accurate identification extremely difficult (7,41).The AC-1, AC-2 and BAC-3 immunofluorescence patterns are presented in Figure 4.The recognition of even isolated anti-DFS70 by IIF is challenging.Samples with mixed patterns containing traditional ANA patterns associated with AC-2 are incorrectly identified even by experienced laboratory technicians (6,7).Recently a novel HEp-2-IIF pattern, similar to AC-2, has been described and named DFS-like or "pseudo-DFS" pattern.This pattern depicts a uniform brightness of the nuclear speckles without the heterogeneity in size and distribution of the speckles seen in the classical AC-2 pattern.The chromatin plate keeps the same features of the nucleoplasm (42).Considering the description, this novel pattern is also very similar to the BAC-3 pattern.Like the AC-4 pattern, maybe an umbrella terminology, bringing together all these patterns, would be the simplest solution to put into practice.Then all patterns morphologic related to the "DFS family" would be identified with the AC code and an identity for the subpattern (e.g., AC-2a, AC-2b…).The accurate interpretation of this "umbrella pattern" would imply confirmation of anti-DFS70 reactivity and other follow-up tests, according to clinical suspicion.The BAC-3 pattern was introduced in the Brazilian decision tree in the IV BCA, which occurred in 2013.Almost a decade later, its importance is recognized by only 23% of Brazilian rheumatologists (43).

Autoantibodies specificities
The antigenic specificity of the anti-cell autoantibodies was tested in 242 (21.8%) samples.A restricted number of specific autoantibodies were available for testing and the majority of the requests comprised basic anti-ENA profile (Ro, La, Sm, U1-RNP) and anti-dsDNA.A total of 56 samples in 242 sera were positive for at least one antibody and only 20 (8.2%) sera presented more than one reactivity.We do not adopt reflex testing or any algorithm for autoantibody testing based on the ANA pattern.The follow-up testing for specific autoantibodies must be ordered by the clinician (44).
Anti-dsDNA, anti-Ro/SSA anti-La/SSB, and anti-Sm were the most requested antibodies, but the most prevalent were anti-Ro/ SSA, anti-Sm, and anti-La.In general, anti-ENA antibodies are identified as the most prevalent in ANA-positive patients.There are some reports in the literature of higher prevalence of anti-Ro (3.9%) (17), anti-DFS70 (3.2%), and anti-nDNA (1.7%) (45) in ANApositive individuals from population-based studies.In patientbased studies the positivity rate varies from 19-44% depending on the diseases included, the methodology of testing, and the number of autoantibodies tested.Anti-ENA and anti-nDNA are routinely included in the requests.Among these autoantibodies, the anti-Ro is usually the most prevalent antibody detected (10,46,47).
In this study, we analyzed whether the reported combined patterns corresponded to those expected considering the autoantibodies identified.The simultaneous presence of two or more autoantibodies was detected in only 20 samples.We observed that in 6 of these samples, one or more expected patterns were not correctly identified.All of them were classified as mixed, according to the VI BCA proposal.According to BCA, the mixed subgroup of CPs comprises a mixture of "different patterns in the same cellular domain, not readily and clearly identified at visual reading".For these cases, the BCA recommends reporting the description "nuclear mixed pattern" without the AC codes (1).We did not find any reference about the positivity of specific antibodies in samples depicting mixed patterns to make a comparison with our findings.
Mitotic cell analysis and sample dilution can help in the pattern identification process.Metaphase plate staining characteristics facilitate the recognition of certain ICAP/BCA codes.Patterns such as AC-1, AC-2, AC-3, BAC-1, BAC-3, and AC-29 present a staining pattern in the metaphase plate that helps to distinguish them from each other, from the AC-4 pattern and from other patterns where the mitotic chromatin is not stained.Sample dilution procedures allow the distinction between overlapping patterns that are present in different titrations (39).In our analysis, only 204 (20.3%) samples in the multiple and mixed subgroups (n=1,005) showed differences in the pattern's titers.We believe this number could be higher.The presence of one pattern in moderate or high titles may have discouraged the search for an end-titer definition of the other pattern.The added value for the end-titer definition of extremely high titer samples is still a matter of debate; however, end-point titration is recommended by international organizations and the BCA, especially in situations of overlapping patterns (39,48).In samples with discordant titers, the fluorescence intensity of the discrete nuclear dots (AC-6,7), centromere (AC-3), and nucleolar pattern (AC-8,9,10) prevailed over the other patterns.

Study limitations
The retrospective design of the study may have had an impact on the number of CPs classified as mixed since it is difficult to judge retrospectively if and how much the pattern presented in the combination observed was different from the expected original fluorescence pattern.Retrospection may also have influenced the number of samples with discrepant titles, which was likely underestimated in our study.The lack of clinical data did not allow the CPs of clinical relevance analysis.

Final considerations
ICAP and BCA recommendations promote considerable improvements in the daily routine workup of autoantibodies testing on HEp-2 cells.The benefits of the nomenclature standardization process are recognized worldwide.All patterns have a nominal, numerical, and group identity, which appears to be sufficiently informative for the decision-making process in the clinical setting.The complex pattern proposal brings relevant concepts to the reading and interpretation routine of the HEp-2-IIF testing.These concepts undoubtedly will favor the analysis and interpretation of fluorescence patterns in the HEp-2 cells, promoting improvements in the analytical phase of the test.The inclusion of information about subtypes of complex patterns in the reporting phase of the HEp-2-IIF assay probably will not add value to the diagnostic decision-making process.

Conclusions
Almost 10% of positive results in routine HEp-2-IIF testing displayed a combination of fluorescence patterns classified as CPs.CPs as a group were among the fourth most frequently encountered patterns in the HEp-2-IFA laboratory routine.Considering the HEp-2 cell anatomy, the double nuclear pattern was by far the most observed cellular domain combination.Among the 3 subtypes of CPs proposed, the multiple pattern was the most prevalent, especially in the pediatric population.The AC-4, AC-2, and AC-6,7 were the most prevalent single patterns observed in the combinations described in this study.There was a significant association between age and the prevalence of most pattern combinations identified.The AC-4+AC-6,7 combination was the most prevalent complex pattern detected regardless of the age group.The concepts involved in the CPs subtype definition should add value to the reading and interpretation of the HEp-2-IIF assay.

4 HEp- 2
FIGURE 4 HEp-2 Patterns -Nuclear Homogeneous and Speckled patterns with stained chromatin plate.(A) AC-1 -Nuclear Homogeneous: uniform diffuse fluorescence covering the entire nucleus.The chromatin plate of metaphase cells depicts a smooth, hyaline, fluorescence pattern.The nucleoli are often covered by the nuclear fluorescence.The anti-dsDNA antibodies, carried out by ELISA, were positive.(B) BAC-3 -Nuclear Speckled Quasi-Homogeneous: Speckled fluorescence, approaching the homogeneous texture throughout the nucleus.The chromatin plate depicts the same pattern.The nucleoli are not stained.The anti-dsDNA antibodies, carried out by ELISA, were negative.(C) AC-2 -Nuclear Dense Fine Specked: Fine speckles throughout the interphase nucleus, with heterogeneity in size, and brightness.Denser and looser areas of speckles are seen throughout the nucleus.The chromatin plate is strongly stained.The nucleoli are not stained.The anti-DFS70/LEDGF antibodies, carried out by ELISA, were positive.

TABLE 1
Stained cellular domains in 1,005 serum samples depicting multiple and mixed patterns on HEp-2 cells indirect immunofluorescence assay (HEp-2-IFA).

TABLE 2
Frequency (%) of the most prevalent ICAP/BCA single patterns and the complex patterns as a group in 11,478 HEp-2-IFA positive samples, according to two age groups.

TABLE 4
Single and complex patterns on HEp-2 cells, according to cellular domains reported by different authors.
*Nu, Nuclear; Ncl, Nucleolar; Cy, Cytoplasmic; Mi, Mitotic.✦ Mixed patterns are mentioned by the authors but not specified.✧ Krzemien et al., 2022 (13) the samples analyzed by the authors comprised 1,731 sera presenting with ANA patterns considered "rare" by the researchers.The percentage 50% is related to the "rare" samples.The number (%) of mixed patterns considering all the ANA-HEp-2-positive samples is not possible to define in the article.

TABLE 3
Frequency (%) of the most prevalent Complex Patterns in 1,111 consecutive HEp-2-IFA positive samples, according to two age groups.