This retrospective study was based on available data at the Pathology Department of Shohada-e Tajrish Hospital in the period from January 2014 to April 2018. Only scalp lesions that had both clinical and histopathological criteria for a diagnosis of scarring alopecia consistent with “discoid lupus erythematosus” or “lichen planopilaris” were included. For all the patients, the histopathological diagnosis was based on H&E stained sections and 3 special stainings, including PAS (Periodic Acid Schiff) for the evaluation of basement membrane thickening, alcian blue at pH 2.5 for demonstrating the presence and pattern of dermal mucin deposition and orcein staining for the evaluation of elastic fibers. The histopathological criteria for DLE were comprised of interface dermatitis at the interfollicular epidermis and the perifollicular region, dense perivascular and periadnexal lymphoid infiltration (both superficial and deep dermal), increased dermal mucin deposition, and thickening of the basement membrane. The histological criteria for LPP were perifollicular interface dermatitis at the level of the infundibulum and the isthmus and a lack of involvement at the inferior segment of the hair follicle and interfollicular epidermis. Overall, 24 cases of DLE and 30 cases of LPP were included. All the cases were categorized into 2 stages according to the amount of inflammation and fibrosis: the early/active stage, characterized by lymphocytic infiltration and the absence of fibrosis, and the late stage, which showed destroyed hair follicles and replacement by fibrous tracts and minimal or absent lymphocytic infiltration.19,20

Formalin-fixed paraffin-embedded tissue sections from all the cases were immunostained for the CD123 marker, mouse anti-human CD123 (Interleukin-3 Receptor Alpha Chain) monoclonal antibody (Clone 7G3) (Master Diagnostica, Spain). From the paraffin blocks, 4μm (micrometer) tissue sections were prepared on charged slides and then incubated overnight at 37°C. The sections were deparaffinized, rehydrated, and exposed to epitope retrieval by heat (boiling tissue in EDTA buffer at pH 9 for 30min at 95°C). The process was followed by rinsing in distilled water for 3–5 changes and cooling at room temperature for 20min. Endogenous peroxidase was blocked using a peroxidase solution for 10min at room temperature. The primary antibody (which was ready to use) was incubated for 30min, followed byEnVision for 30min at 37°C and rinsing with phosphate-buffered saline and the IHC wash buffer. Thereafter, the antibody was detected by DAB (3,3′-diaminobenzidine) for 2min and was, then, washed with distilled water. Slide mounting was carried out after hematoxylin counterstaining.

Three parameters of PDCs were mainly evaluated on IHC-stained sections (Table 1):

• (1)

  The percentage of PDCs (CD 123+), calculated by dividing the PDC count by the mononuclear cell count at 3 separate regions that had the most prominent infiltration in serial sections. Then, the PDC content was defined as score 1 (PDC<10%), score 2 (10%≤PDC<20%), and score 3 (PDC≥20%) (18).

• (2)

  The predominant anatomic location of the PDC infiltrate, including the perifollicular, perivascular, perieccrine, subcutaneous, intrafollicular, dermoepidermal junction, and intraepidermal regions.

• (3)

  The presence of the PDC cluster, defined as the aggregation of at least 5 PDCs. Aggregations of 5–10 PDCs were considered as small clusters and aggregations of more than 10 PDCs were considered as large clusters.17,18 Additionally, the anatomic location of small and large clusters was determined according to the locations described above.

The percentage of PDCs was compared between the 2 groups using the independent sample t-test. The distribution patterns of PDCs were compared between DLE and LPP using the Fisher exact test. The presence of clusters was compared between the 2 groups using the χ2 test. A p-value of less than 0.05 was considered statistically significant.

*All procedures performed in this study involving human participants were in accordance with the ethical standards of our institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

This study was approved by our institutional research committee with no. “IR.SBMU.SRC.REC.1395.43”.

In total, 54 IHC-stained slides were examined. DLE specimens were taken from 24 patients (7 males and 17 females) at a mean age of 45.33±12.47 years, and LPP specimens were taken from 30 patients (12 males and 18 females) at a mean age of 49.50±13.52 years. There was no significant difference between the 2 groups regarding the age (p=0.245). Nine (37.5%) cases of DLE and 15 (50%) cases of LPP were in the early/active stage of disease, and 15 (62.5%) cases of DLE and 15 (50%) cases of LPP were in the late stage of disease. There was no significant difference between the 2 groups regarding the stage of disease (p=0.3580).

The mean percentage of PDCs was 14.87±7.99 (range: 4%–34%) in the LPP cases and 25.45±10.78 (range: 10%–52%) in the DLE cases, which showed a significant difference between the 2 groups (p<0.001) (Fig. 1). In the DLE cases, 0/24 (0%), 6/24 (25%), and 18/24 (75%) of the specimens were scored 1, 2 and 3, respectively. For the LPP cases, 9/30 (30%), 14/30 (46.6%), and 7/30 (23.3%) of the specimens were scored 1, 2 and 3, correspondingly, which was statistically significant (p<0.001) (Table 1).

Figure 1.

Percentage of plasmacytoid dendritic cells in discoid lupus erythematosus (DLE) and lichen planopilaris (LPP). Middle point, median; Box, interquartile range; Whisker, range (excluding outliers).

(0.04MB).

The perifollicular region was the most common location involved (all the cases of DLE and LPP had perifollicular infiltrations) (Fig. 2). Perivascular involvement was observed in 29/30 (96.7%) of the LPP cases and 22/24 (91.7%) of the DLE cases, which was not statistically significant (p=0.579). The infiltration of the dermoepidermal junction was detected in 15/24 (62.5%) of the DLE cases and 2/30 (6.7%) of the LPP cases, which was statistically significant (p<0.001). Subcutaneous involvement was identified in 11/24 (45.8%) of the DLE cases and 6/30 (20%) of the LPP cases; this differentiation was statistically significant (p=0.042). Perieccrine infiltration was observed in 4/24 (16.7%) of the DLE cases and 0/30 of the LPP cases, showing statistical significance (p=0.034). Interstitial infiltration was observed in 14/30 (46.7%) of the LPP cases and 2/24 (8.3%) of the DLE cases, which was also statistically significant (p=0.02). Intrafollicular involvement was detected in 3/24 (12.5%) of the DLE cases and 0/30 of the LPP cases (Fig. 2), which was not statistically significant. Intraepidermal infiltration of PDCs was observed neither in the DLE cases nor in the LPP cases (Fig. 3 shows the anatomical distribution pattern of PDCs).

Figure 2.

Representative discoid lupus erythematosus case: CD123 immunohistochemistry presence of PDCs at the intrafollicular and perifollicular area (40×).

(0.15MB).

Figure 3.

Distribution patterns of plasmacytoid dendritic cells (PDCs) in lichen planopilaris (LPP) and discoid lupus erythematosus (DLE).

(0.11MB).

Fifteen small clusters (aggregation of 5–10 cells) were identified. Ten of these small clusters were observed in 10 cases of LPP (33.33%) and 5 of them were observed in 4 cases of DLE (16.7%). The location of these clusters in both LPP and DLE was perifollicular and perivascular. Twenty-six large clusters (aggregation of 10 cells or more) were identified. Three of these large clusters were detected in 2 (6.7%) cases of LPP and 23 of them were identified in 12 (50%) cases of DLE. There was a significant difference between these 2 groups regarding cluster formation (p=0.002). The presence of large clusters had 50% sensitivity and 94% specificity for the differentiation of DLE from LPP (Fig. 4A and B).

Figure 4.

Representative lichen planopilaris: presence of PDCs (no cluster formation), 20× (A) and representative discoid lupus erythematosus case: large cluster (aggregation of 10 cells or more) of plasmacytoid dendritic cells, 40× (B).

(0.28MB).

The anatomic sites of the large clusters were also identified. The large clusters in the DLE specimens were often observed at the perifollicular region (21/23 of the clusters). One large cluster was observed at the dermal–epidermal junction region, and another one was observed at the subcutis. In the LPP specimens, the large clusters were observed at the perifollicular (1), perivascular (1), and interstitial (1) regions.

Half of the LPP cases with small clusters were in the early/active stage of disease and the other half were in the late stage of disease, which was not statistically significant (p=1.000). Of all the DLE cases with small clusters, one case had active disease and 3 cases were at the late stage of disease, which was not statistically significant (p=1.000). Two cases of LPP had large clusters: 1 in the active stage of disease and the other one in the late stage of disease, which was not statistically significant (p=1.000). Of the 12 cases of DLE with large clusters, 4 cases had active disease and 8 cases were at the late stage of disease, which was not statistically significant (p=0.673).