Profile of immunological markers in skin biopsies from patients with probable and confirmed systemic lupus erythematosus

Full Text

INTRODUCTION

The detection of immunoreactants, including IgA, IgM, IgG, C3, and C1q, by direct immunofluorescence in skin biopsy specimens from patients with systemic lupus erythematosus (SLE), is known as the lupus band test (LBT) [1–3]. The test was originally used to differentiate skin lesions in SLE and discoid lupus erythematosus. As a rule, immunoreactants were not found in a biopsy specimen of an unaffected skin area in patients with discoid SLE [4, 5]. However, further studies revealed a high incidence of immunoreactants (50%–77%) in biopsies of unaffected skin in patients with SLE [6–8]. In addition, the authors established certain correlations between immunoreactants in the LBT and SLE activity and ongoing therapy [6, 8–11].

LBT assessment in patients with probable SLE when patients have specific immunological markers and some clinical symptoms similar to SLE, but they do not fully meet the classification criteria of the American Rheumatism Association or Systemic Lupus International Collaborating Clinics (SLICC 2012) is of particular interest [8, 12–14]. Probable SLE may develop over time into definitive SLE or other connective tissue diseases, such as Sjogren disease and rheumatoid arthritis. However, in some cases, it can remain at the immunological disorder stage for a long time and not transform into certain nosological forms [12, 14–16].

In the literature, only a few data are presented on LBT values of probable SLE. Determining an immunoreactant profile by assessing the nature, quantity, degree of immunofluorescence, and site of their detection in the skin biopsy are of both scientific and practical interest.

MATERIALS AND METHODS

This study included 94 patients with an elevated antinuclear factor titer. Of these, 56 patients were diagnosed with significant SLE since they had at least four SLICC 2012 classification criteria. The group with probable SLE included 38 patients who had less than four SLICC 2012 criteria.

The study did not include patients with SLE under 18 years, pregnant and breastfeeding women, and persons with severe concomitant pathology of the cardiovascular system, gastrointestinal tract, liver, kidneys, and malignant neoplasms. The study was approved by the local committee of the North–Western State Medical University named after I.I. Mechnikov.

At the next stage of the study, patients with significant SLE were divided into two subgroups. The first subgroup consisted of 25 patients with advanced SLE who received glucocorticoid and disease-modifying anti-rheumatic drugs, and the second subgroup included 31 patients with newly diagnosed (early) SLE.

All patients underwent clinical, laboratory, and instrumental examinations using routine diagnostic methods. Disease activity was assessed using the Systemic Lupus Erythematosus Disease Activity Index modified 2K (SLEDAI–2K). Antibodies against double–stranded DNA were determined by enzyme immunoassay (reference values–0–25 U/ml). Along with this, a biopsy of intact skin in the outer surface area of the upper third of the shoulder (in the area of the deltoid muscle) was performed. Immunoglobulins IgA, IgM, and IgG, and complement components C3, C1q (LBT) were detected in skin biopsy using immunohistochemical methods in the laboratory of the Scientific and Methodological Center for Molecular Medicine of the First Pavlov Saint-Petersburg State Medical University. If immunoreactants were detected in the biopsy specimen, the LBT was considered positive. LBT was assessed based on the type of IgA/IgM/IgG/C1q–/C3–complement components deposits, the intensity of their luminescence (+ to +++), the nature of the luminescence (fine–granular, granular, linear), the site of immunoreactant deposition (small vessels of the dermis; papillary layer of the dermis; basement membrane zone–dermoepidermal junction).

Statistical processing of the results was performed using the statistical data analysis package Statistica 10.0 for Windows (Statsoft Inc., USA), including parametric and nonparametric analysis methods. Differences were considered statistically significant at p < 0.05.

STUDY RESULTS

In subjects with significant SLE (n = 56), 86% were women; their mean age was 37.3 years, and the disease duration was 2.5 years. The disease activity according to the SLEDAI–2K index corresponded on average to 9.2 ± 5.9 points. In addition, an increased level of antibodies against double–stranded DNA (dsDNA) was observed in 25 (73.5%) patients. In the subjects with probable SLE (n = 38), 92.1% were women with a mean disease duration of nine months, and the mean age was 38.1 years. The disease activity was significantly lower than in the group with significant SLE and corresponded to 4.3 ± 2.4 points. Also, increased antibody levels against dsDNA were observed in 10 (26.3%) patients.

The data analysis showed that positive LBT was detected in 34 (60.7%) patients with significant SLE and 18 (47%) patients with probable SLE. Analysis of the study results in the patient subgroups with confirmed SLE showed that LBT was positive in 23 (74%) patients with early SLE. In addition, only 11 (44%) patients had an advanced clinical and laboratory form of the disease (Fig. 1).

 

Fig. 1. Lupus band test (LBT) results in the studied patient groups. SLE — systemic lupus erythematosus

Рис. 1. Результаты теста волчаночной полоски (ТВП) у пациентов исследуемых групп. СКВ — системная красная волчанка

 

Thus, an LBT positive result was significantly more often recorded in the group with early SLE than the group with advanced SLE (p = 0.02) and probable SLE (p = 0.02). The lower detectability of immunoreactants in skin biopsies in patients with advanced SLE may be related to pathogenetic therapy in this group [84% of patients received oral glucocorticoids at low and moderate doses, 36% of subjects received combined pulse therapy with high doses of methylprednisolone and cyclophosphamide, 40% of subjects received cytostatic therapy (methotrexate, azathioprine, or mycophenolate mofetil), 64% of subjects received hydroxychloroquine (Plaquenil), 24% of subjects received non–steroidal anti–inflammatory drugs, and 8% of subjects received the genetically engineered biological drug rituximab].

The frequency, nature, and various combinations of specific immunoreactants detected in all groups are shown in Table 1. More than one immunoreactant (IgA, IgM, IgG, C3, C1q) was found in all groups, including 66.7% of cases with probable SLE. The most common immunoreactant was IgM, and various combinations (IgM + IgG) were recorded significantly more often in the group of significant (61.8%). Advanced SLE (72.7%) compared with probable SLE (27.7%) (p = 0.02 and p = 0.018, respectively). When comparing the groups of patients with early and probable SLE, no significant differences were obtained (p = 0.066). The combination of three immunoreactants (IgG + IgM + C3) was detected significantly more often in the groups of examined patients with significant, extensive, and early SLE compared with probable SLE, where such a combination was found in only 5.5% of cases (p = 0.03, p = 0.03, and p = 0.046, respectively). In the study of other combinations of immunoreactants in the compared groups, no significant differences were established.

 

Table 1. The profile of immunoreactants deposited in the skin biopsy materials of patients with LBT (+)

Таблица 1. Профиль выявляемых иммунореактантов в кожных биоптатах у обследованных пациентов c положительным тестом волчаночной полосы

Immunohistochemical data	Systemic lupus erythematosus
	expanded (n = 11)	early (n = 23)	significant (n = 34)	probable (n = 18)
IgG	8 (72.7)	14 (60.9)	22 (64.7)	8 (44.4)
IgM	10 (90.9)	19 (82.6)	29 (85.3)	13 (72.2)
IgA	3 (27.3)	3 (13)	6 (17.6)	7 (39)
C3	7 (63.6)	13 (56.5)	20 (58.8)	8 (44.4)
С1q	4 (36.4)	9 (39.1)	13 (38.2)	7 (39)
One immunoreactant
	1 (9.1)	5 (21.7)	6 (17.6)	6 (33.3)
One immunoreactant
	10 (90.9)	18 (78.3)	28 (82.4)	12 (66.7)
Two immunoreactants
IgG + IgM	8 (72.7)*	13 (56.5)	21 (61.8)*	5 (27.7)*
Three immunoreactants
IgG + IgM + C3	4 (36.4)*	7 (30.4)*	11 (32.4)*	1 (5.5)*

* p < 0.05.

 

Although positive LBT was more often recorded in patients with early SLE compared with advanced SLE (74% and 44%, respectively, p = 0.02) when assessing the nature and incidence of specific immunoreactants in the skin biopsy, the data of these two groups were comparable. In early SLE, IgG and IgM were determined more often (60.9% and 82.6%, respectively), and in probable SLE–IgM (72.2% of cases).

The data obtained served as the basis for studying the peculiarities of immunoreactant distribution in biopsy of unaffected skin in patients of the compared groups considering the localization, intensity, and nature of their distribution (Table 2). In advanced SLE, immunoreactants IgG (72.7%), IgM (90.9%), and C3 (63.6%) with varying degrees of luminescence intensity were mainly found. In contrast, in patients with early SLE luminescence of IgG (p = 0.01) and IgM (p = 0.004) of moderate intensity was detected significantly more often than in subjects with probable SLE. Also, deposits of immunoreactants were detected in small vessels of the dermis that may indirectly indicate the presence of immunocomplex vasculitis.

 

Table 2. Character, location, and degree of fluorescence immunoreactants in the groups of patients with LBT (+), n (%)

Таблица 2. Характер, локализация и интенсивность свечения иммунореактантов в обследованных группах пациентов с положительным тестом волчаночной полосы, n (%)

Parameters	Systemic lupus erythematosus
	expanded (n = 11)	early (n = 23)	probable (n = 18)
IgA, glow:
weak (1+)	3 (27.3%)	1 (4.3%)	4 (22%)
moderate (2+)	0 (0)	1 (4.3%)	0 (0)
intense (3+)	0 (0)	1 (4.3%)	3 (16.7%)
IgG, glow:
weak	4 (36.4%)	4 (17.4%)	6 (33.3%)
moderate	3 (27.3%)	9 (39.1%)*	1 (5.6%)*
intense	1 (9.1%)	1 (4.3%)	1 (5.6%)
IgM, glow:
weak	3 (27.3%)	3 (13%)	6 (33.3%)
moderate	3 (27.3%)	14 (61%)*	3 (16.7%)*
intense	4 (36.4%)	2 (8.7%)	3 (16.7%)
C3 component of complement, glow:
weak	3 (27.3%)	6 (26%)	5 (27.8%)
moderate	3 (27.3%)	7 (30.4%)	2 (11.1%)
intense	1 (9.1%)	0 (0)	1 (5.6%)
C1q component of complement, glow:
weak	3 (27.3%)	7 (30.4%)	4 (22.2%)
moderate	1 (9.1%)	2 (8.7%)	3 (16.7%)
intense	0 (0)	0 (0)	0 (0)
The nature of the deposits:
fine–grained	6 (54.5%)	17 (74%)	9 (50%)
granular	5 (45.5%)	6 (26%)	7 (39%)
linear	0 (0)	0 (0)	2 (11.1%)
Localization of immunoreactants:
in small vessels of the dermis	2 (18.2%)	7 (30.4%)	5 (27.8%)
in the papillary layer of the dermis	1 (9.1%)	2 (8.7%)	1 (5.6%)
along the basement membrane of the epidermis	8 (73%)	14 (61%)	12 (66.7%)

* p < 0.05.

 

In the LBT results analysis, depending on the activity of SLE and the level of antibodies against dsDNA, a relationship between positive LBT, the activity of SLEDAI–2K disease, and the antibody titer to dsDNA were established. Thus, SLE activity in patients with LBT (+) was 10 points, and in patients with LBT (−), it was 5 points (p = 0.001) (Fig. 2). At the same time, the titer of antibodies to dsDNA in patients with SLE who had LBT (+) amounted to 73.7 [24.7; 210] U/ml, and with LBT (−)—only 15.9 [4.1; 38.2] U/ml (p = 0.001) (Fig. 3).

 

Fig. 2. SLEDAI 2K (Me [25%; 75%]) groups of LBT (+) and LBT (–) of the patients with systemic lupus erythematosus (n = 56)

Рис. 2. Активность системной красной волчанки по данным SLEDAI-2K (Me [25 %; 75 %]) в группах с положительным [ТВП (+)] и отрицательным [ТВП (–)] тестом волчаночной полосы у пациентов с достоверной системной красной волчанкой (n = 56)

 

Fig. 3. Level of antiDNA (Me [25%; 75%]) in the groups of LBT (+) and LBT (–) of the patients with systemic lupus erythematosus (n = 56)

Рис. 3. Титр антител к дсДНК (Me [25 %; 75 %]) у пациентов с достоверной системной красной волчанкой (n = 56) в группах с положительным [ТВП (+)] и отрицательным [ТВП (–)] тестом волчаночной полосы

 

The most common immunoreactant in patients with significant SLE was IgM (29%–85.3%), whereas an increase in the level of antibodies against dsDNA was noted in 25 (73.5%) patients. In the presence of IgM and its various combinations with other immunoglobulin classes and complement components, an increase in the titer of antibodies against dsDNA was significantly more often detected (p < 0.05) (Table 3). At the same time, a higher level of dsDNA was found in patients with IgM + IgA association (the median of antibodies against dsDNA was 179.15 U/ml) (Fig. 4). Along with this, a relationship between an increase in the level of antibodies to dsDNA and a moderate (p < 0.05) or high (p < 0.05) degree of IgM immunofluorescence was established (Table 4).

 

Fig. 4. The level of antiDNA (Me [25%; 75%]) depending on different associations of immunoreactants in the skin biopsy materials of the patients with systemic lupus erythematosus (n = 56)

Рис. 4. Уровень антител к дсДНК (Me [25 %; 75 %]) в зависимости от различных ассоциаций иммунореактантов в биоптате кожи у пациентов с достоверной системной красной волчанкой (n = 56)

 

Table 3. The correlation between different combinations of immunoreactants and increased antiDNA in patients with systemic lupus erythematosus (n = 56)

Таблица 3. Корреляционные связи между различными сочетаниями иммунореактантов и повышением титра антиДНК у пациентов с достоверной системной красной волчанкой (n = 56)

Parameter	Lupus band test result, n (%)
	LBT (−)	IgM + IgA	IgM + C3	IgM + C1q	IgM + IgG	IgM + IgG + C3	IgM	C1q	IgM + C3 + C1q
	n = 22	n = 6	n = 17	n = 10	n = 21	n = 11	n = 29	n = 13	n = 7
AntiDNA >25 U/ml	7 (31.8)	4 (66.7)	14 (82.4)*	9 (90)*	14 (66.7)*	8 (72.7)*	22 (75.9)*	11 (84.6)*	6 (85.7)*

Note. LBT (−) – negative lupus strip test. * p < 0.05 compared with the LBT group (−).

 

Table 4. Relationship between the degree of IgM immunofluorescence and the increase in antiDNA of patients with systemic lupus erythematosus (n = 56)

Таблица 4. Взаимосвязь между степенью иммунофлюоресценции IgM и повышением антител к двухспиральной ДНК у пациентов с достоверной системной красной волчанкой (n = 56)

Parameter	Immunofluorescence IgM, n ( %)
	LBT (−) n = 22	1+ (weak) n = 6	2+ (moderate) n = 17	3+ (intense) n = 6
AntiDNA >25 U/ml	7 (31.8)	4 (66.7)	13 (76.5)*	5 (83.3)*

Note. LBT (−) – negative lupus strip test. * p < 0.05 compared with the LBT group (−).

 

DISCUSSION

This study revealed a high incidence of immunoreactants (60.7%) in patients with SLE, confirmed by data from several authors. For example, Zecevic (2001) found immunoreactants in 23 (60.5%) of 38 patients with SLE [7], and Gangaram et al. (2004)–in 63% of patients with SLE on unaffected photo-exposed areas of the skin [6]. In patients with advanced SLE, the frequency of a positive test was lower and amounted to 44%, which could be explained by the effect of combined therapy on the studied immunoreactant expression. However, the possibility of the therapy effect on LBT parameters is still under discussion. Thus, according to Davis and Gilliam (1984), the initial LBT (+) or LBT (−) more often remained unchanged at repeated skin biopsy in patients with SLE during treatment with prednisolone (daily dose–40 mg) or cytostatic drugs [9]. According to other authors, there was a direct relationship between immunoreactant levels and SLE activity [10, 17]. Moreover, Provost et al. (1980) showed that immunoreactants might disappear, or their number may decrease. The intensity of their fluorescence may decrease with a decrease in SLE activity or achievement of remission [18].

In our study, a relationship between positive LBT and disease activity was established according to SLEDAI–2K data (p = 0.001) as well as an increased titer of antibodies against dsDNA (p = 0.001) compared with LBT (−). Furthermore, in the study performed by Gangaram and Kong (2004), a significant correlation was also noted between LBT (+) in the skin and antibodies against dsDNA (p = 0.02). However, the authors did not observe the relationship between LBT, disease activity, and the presence of kidney damage and skin lesions. This served as the basis for the assertion that LBT can be used to diagnose SLE but cannot assess disease activity [6].

The most frequently detected immunoreactant in patients with SLE with LBT is IgM (approximately 90% of cases), IgA is much less common [19, 20]. Our study also confirms these data since IgM in skin biopsies of patients with significant SLE was determined in 85.3% of cases, and IgA–only in 17.6%. Such a high detection rate of IgM can be associated with its larger size compared with other proteins. Therefore, it can remain in the dermoepidermal junction for a longer time than other immunoreactant representatives [5, 18]. The deposition of immunoglobulins and complement components in the skin of patients with SLE is a dynamic process. It can vary depending on various factors (environment, stress, infections, drugs, others) [5, 18].

Previously, most researchers noted that with a higher activity of SLE and in the presence of renal damage and high titers of antibodies against dsDNA, IgG was more often detected in skin biopsies. In contrast, IgM deposits were associated with a more favorable disease course without kidney damage [11, 18]. Detection of weak immunofluorescence of monoimmunoreactant IgM in skin biopsy is less specific for SLE, and weak intermittent immunofluorescence of IgM and C1q deposits in the dermoepidermal junction can be detected, including in healthy individuals, in almost 20% of cases [21, 22]. In contrast, Permin et al. (1979) showed that in skin biopsies of 500 patients with SLE and other diseases, LBT was positive in ³/₄ of patients with SLE, and the deposition of the C1q component of complement was observed mainly in SLE and never in patients with diabetes mellitus, allergic diseases. SLE-like drug-induced syndrome indicates the specificity of the C1q immunoreactant for SLE [23]. In a study performed by Leibowitch et al. (1981), C1q deposits were found in 90% of patients with SLE and only 29% of patients with discoid lupus erythematosus. This finding led the authors to conclude that C1q deposits in the skin may be a valuable SLE marker [24].

In the study performed by Minz et al. (2010) in SLE, the immunoreactant IgM was most often detected (85%), its combination with IgG was noted in 77% of cases, and IgM in combination with IgG and C3 was observed in 46% of patients [25]. On the other hand, Luo (2013) reported that in patients with SLE, IgM was detected in a skin biopsy in 86% of cases, and a combination of IgM with C3 was found in 28% of patients. At the same time, the presence of several immunoreactants simultaneously correlated with disease activity, whereas the detection of only one immunoreactant was not very informative for predicting SLE activity [11].

In our study, isolated immunoglobulins G and M in skin biopsies from patients with a confirmed diagnosis of SLE practically did not occur; IgG was combined with other classes of immunoglobulins and complement components. At the same time, the combination of IgG + IgM was detected in 61.8% of patients with a significant SLE diagnosis, and the combination of IgG + IgM + C3 was recorded in 31.4%. The incidence of C1q was 38.2%.

The assessment of LBT in patients with probable SLE is of particular interest. Several studies are presented in the literature. According to Ullman (1975), although patients had antinuclear antibodies in serum and symptoms similar to SLE, they were not enough to confirm the diagnosis. However, immunoreactants in biopsies of unaffected skin were found in 1/3 of patients [26]. Akarsu et al. (2017) studied LBT in patients with borderline SLE (the presence of antinuclear antibodies and mucocutaneous manifestations) and SLE with discoid lesions. They found a similarity to SLE in deposits in skin biopsies of IgM and IgG. However, they were less likely to have deposits of multiple conjugates than SLE [27]. According to Goldstein et al. (1985), when examining 33 patients with undifferentiated connective tissue disease (including probable SLE), the LBT results did not differ significantly. However, 18% of patients from this group subsequently developed significant SLE, and 6% had rheumatoid arthritis [28]. The results of a prospective study performed by Leibowitch et al. (1981) showed that of 42 patients with discoid SLE, four (9.5%) patients who had C1q in the skin biopsy transformed into SLE. The authors suggested that the presence of C1q in the skin biopsy may be directly related to the risk of developing systemic autoimmune diseases [24].

In our study, the immunoreactant detection rate in biopsy of unaffected skin in patients with probable SLE was 47%. The most frequently detected immunoreactant was IgM (72.2%). However, in 16.6% of cases, it was found in isolation, its combination with IgG was observed in five (27.7%) patients, and the combination of IgM + IgG + C3 was observed in only one (5.5%) patient. Immunoreactants M and G in this group of subjects were characterized by weak immunofluorescence. In seven (39%) patients with probable SLE, C1q was detected in skin biopsies. How many of them will transfer to the group with significant SLE can be answered during follow-up.

CONCLUSION

The analysis of LBT results in patients with probable SLE showed some similarity with significant SLE (positive LBT in 47% of cases, IgM detection in 72.2% of patients). However, in 33.3% of patients with probable SLE, an isolated deposition of a single immunoreactant was found in the skin biopsy, whereas SLE is characterized by a combination of several immunoreactants (IgM + IgG; IgM + IgG + C3).

Thus, considering the detected immunoreactants, their associations, the degree of immunofluorescence, and the site of deposition in the skin biopsy, LBT can be deemed an additional test for verifying the diagnosis of probable and significant SLE.

About the authors

Viktoria A. Lila

North-Western State Medical University named after I.I. Mechnikov

Author for correspondence.
Email: liu_lo@mail.ru
ORCID iD: 0000-0001-5006-3358
SPIN-code: 8348-2910

PhD student

Russian Federation, 41 Kirochnaya str., Saint Petersburg, 191015

Vadim I. Mazurov

North-Western State Medical University named after I.I. Mechnikov

Email: maz.nwgmu@yandex.ru
ORCID iD: 0000-0002-0797-2051
SPIN-code: 6823-5482
Scopus Author ID: 16936315400
ResearcherId: A-8944-2016

MD, Dr. Sci. (Med.), Professor, Honoured Science Worker, Academician of the RAS

Russian Federation, 41 Kirochnaya str., Saint Petersburg, 191015

References

Supplementary files