Cutaneous lymphoid hyperplasia: a lymphoproliferative continuum with lymphomatous potential

Cutaneous Lymphoid Hyperplasia: A Lymphoproliferative Continuum With Lymphomatous Potential MINAKSHI NIHAL, PHD, DEBRA MIKKOLA, MS, NANCY HORVATH, BSN, ANITA C. GILLIAM, MD, PHD, SETH R. STEVENS, MD, TIMOTHY P. SPIRO, MD, KEVIN D. COOPER, MD, AND GARY S. WOOD, MD Cutaneous lymphoid hyperplasia (CLH) has been proposed to be the benign end of a continuum of lymphoproliferative disorders with cutaneous lymphoma at its malignant extreme. An intermediate condition, known as “clonal CLH,” was first recognized by us and shown to be a transitional state capable of eventuating in overt lymphoma. To better determine the prevalence of dominant clonality and risk of lymphoma among CLH cases, we studied the immunohistology and clonality of fresh-frozen samples from 44 CLH patients referred to a multidisciplinary cutaneous lymphoproliferative disorders program. Using a large panel of lymphoid markers, the cases were divided into 38 typical mixed B-cell/T-cell type CLH and 6 T-cell–rich type (T-CLH), the latter containing > 90% T cells. Of the 44 patients, 38 had solitary or localized lesions (4 cases of T-CLH), and 6 had regional/generalized lesions (2 cases of T-CLH). Forty cases were of idiopathic etiology. Suspected etiologies among 4 other cases included mercuric tattoo pigment, doxepin, clozapine, and bacterial infection. Immunoglobulin heavy chain (IgH) and T-cell receptor (TCR)-gamma gene rearrangements (GR) were studied using polymerase chain reaction assays, which are approximately 80%

sensitive. Overall, 27 cases (61%) showed clonal CLH: 12 IgHⴙ (27%; 3 cases of T-CLH); 13 TCRⴙ (30%; 1 case of T-CLH); and 2 IgHⴙ/TCRⴙ (4%; neither case was T-CLH). Two cases (4%; 1 case of T-CLH) progressed to cutaneous B-cell lymphoma. Both of these patients presented with regional lesions. Our findings indicate that clonal overgrowth is common in CLH, links CLH to lymphoma, and probably involves both B- and T-cell lineages (although TCR GR by B cells and vice versa could not be ruled out). The high prevalence of dominant clonality in our series may have resulted from the sensitivity of our PCR assays as well as patient selection. HUM PATHOL 34: 617-622. © 2003 Elsevier Inc. All rights reserved. Key words: B cells; cutaneous lymphoid hyperplasia; lymphoma; pseudolymphoma; T cells; tumor-infiltrating lymphocytes. Abbreviations: CLH, cutaneous lymphoid hyperplasia; T-CLH, T-cell cutaneous lymphoid hyperplasia; Ig, immunoglobulin; TCR, T-cell receptor; GR, gene rearrangement; PCR, polymerase chain reaction; CBCL, cutaneous B-cell lymphoma; TBE, trisborate EDTA; DGGE, denaturing gradient gel electrophoresis.

Cutaneous lymphoid hyperplasia (CLH) occurs in adults of both sexes and also affects the pediatric age group.1-3 The male-to-female ratio is approximately 1:2 to 1:3. CLH can involve any area of the skin but is most common on the face. The majority of cases present as a solitary or localized cluster of asymptomatic, erythematous to violaceous papules or nodules. Occasionally, the lesions may coalesce into a plaque. Uncommonly, the lesions may be widespread over a body region or even generalized in their cutaneous distribution. In CLH that is not part of some other systemic lymphoid hyperplasia syndrome, constitutional symptoms are absent. No associated extracutaneous physical findings or laboratory abnormalities are seen except in

rare cases associated with regional lymphadenopathy. CLH may be idiopathic, or it may arise in response to a wide variety of foreign antigens, including arthropod bites, stings, and infestations; tattoos; vaccinations; trauma; injection of foreign substances; pierced ear jewelry; and drugs.4-11 Some cases have been attributed to infection with Borrelia burgdorferi.12 These associations suggest that CLH begins as a reactive response to newly encountered antigens. The typical case of CLH exhibits a patchy to confluent dense lymphoid infiltrate throughout the dermis that spares the epidermis and is separated from it by a narrow, so-called Grenz zone. Immunophenotypic studies have shown that most cases of CLH consist of a mixture of reactive polytypic B cells, T cells, macrophages, and dendritic cells.13-15 Occasionally, B cells are rare or absent, and the infiltrate is composed predominantly of T cells. Like nodal lymphoid hyperplasia, CLH lacks the t (14;18) translocation or bcl-2 expression by germinal center cells.15 Interestingly, some cases harbor occult monoclonal B-cell populations as determined by Southern blot or polymerase chain reaction (PCR) analysis of immunoglobulin (Ig) gene rearrangements.15,16 We refer to these cases as “clonal CLH.” Serial molecular biologic studies showed identical Ig gene rearrangements in subsequent cutaneous B-cell lymphoma (CBCL) arising in these same cases, indicating that clonal CLH may undergo clonal evolution to CBCL. These findings support the view that

From the Department of Dermatology, University of Wisconsin and the Middleton Department of Veterans Affairs Medical Center, Madison, WI; the Departments of Dermatology, Pathology, and Medicine (Oncology), and the Skin Diseases Research Center, Case Western Reserve University, Cleveland, OH; and the Louis Stokes Department of Veterans Affairs Medical Center, Cleveland, OH. Accepted for publication December 19, 2002. Supported in part by NIH Grant No. AR02136 (G.S.W.) and merit review funding from the Medical Research Service of Department of Veterans Affairs (G.S.W.). Address correspondence and reprint requests to Gary S. Wood, MD, Department of Dermatology, University of Wisconsin, One South Park, 7th Floor, Madison, WI 53715. © 2003 Elsevier Inc. All rights reserved. 0046-8177/03/3406-0020$30.00/0 doi:10.1016-S0046-8177(03)00075-3

617

HUMAN PATHOLOGY

Volume 34, No. 6 (June 2003)

CLH, clonal CLH, and CBCL are points on a single continuum of B-cell lymphoproliferative disease along which individual patients can progress to increasingly aggressive forms of the disease. In the current study, we used PCR-based analysis of IgH and T-cell receptor (TCR) gene rearrangements to determine the clonality of lesional lymphoid cells in a series of 44 patients with CLH. We were particularly interested in determining the technical suitability of PCR-based clonality assays applied to CLH specimens, the consistency of clonality among serial CLH samples, the prevalence of clonal CLH, and the progression of CLH to CBCL. MATERIAL AND METHODS Patients and Specimens Forty-four patients with CLH were referred to a multidisciplinary cutaneous lymphoproliferative disorders program during the study period (1991 through 2000). All test specimens were cutaneous punch or shave biopsy specimens obtained from well-developed lesions. Representative portions of each specimen were divided for routine histopathology and fresh-frozen immunoperoxidase and molecular analysis.

DNA Extraction DNA was isolated from fresh-frozen skin biopsy specimens, neoplastic B-and T-cell lines, and reactive tonsils from healthy individuals via proteolysis followed by phenol-chloroform extraction.17 The minced tissues or cells were digested overnight at 37°C in an initial suspension solution containing 0.8 ⫻ sodium chloride sodium citric acid, 0.2 mmol/L NaCl, 0.5% sodium dodecyl sulfate, 1mmol/L dithiothreitol (DTT), and 50 ␮L proteinase K (1mg/mL). Double phenolchloroform extraction was performed followed by precipitation of DNA with 95% cold ethanol, subsequent washing with 80% ethanol, vacuum drying, and resuspension of DNA pellets overnight in a solution containing 1mmol/L each of Tris-HCl, NaCl, and ethylene diamene tetraacetic acid.

FR1/FR2 IgH PCR The FR1/FR2 IgH PCR method used a multiplex nested PCR approach. The first round employed a mixture of the consensus primers directed against the framework I (FR1) region (VH 1/5, 2, 3, 4, and 6) and the joining region primer LJH of the Ig gene.18 The second round used the FR2A and VLJH primers to amplify the first-round products. The final concentrations of the reactants for the first round were as follows: 50 mmol/L KCl; 10 mmol/L Tris-Cl at pH 8.3; 200 ␮m each dNTPs; 2.5 mmol/L MgCl2; 300 ng FR I primers (50 ng each VH primer); 75 ng LJH primer; 1 ␮g DNA template; and 2.5 U Taq polymerase. Cycle conditions for the first round were 95°C for 15 seconds, 63°C for 30 seconds, and 72°C for 30 seconds for the first 5 cycles. The remaining 35 cycles were done at 96°C for 15 seconds, 57°C for 30 seconds, and 72°C for 30 seconds. Second-round PCR used similar concentration of all the reactants except for the following changes: 1.5 mmol/L MgCl2 as well as 200 and 400 ng each FR2 and VLJL primers. Of first-round product, 1/100 was used as template in the second round. PCR conditions for this round were 95°C for 15 seconds, 63°C for 30 seconds, and 72°C for 30 seconds for 40 cycles. Approximately 7 ␮L PCR products was electrophoresed through 3% agarose trisborate

EDTA (TBE) gel, followed by visualization of bands with ethidium bromide staining. Expected product size was approximately 240 to 260 base pairs (bp).

FR3 IgH PCR FR3 IgH PCR involves a monoplex one-step PCR approach that used consensus primers for the FR3 portion of the VH and JH regions of the IgH gene.19 The final concentrations of the reactants in a 100-␮L mixture were as follows: 200 ␮m each dNTPs; 50 mmol/L KCl; 10 mmol/L Tris-Cl at pH 9.0; 1.5 mmol/L MgCl2; 100 ␮g/mL bovine serum albumin (BSA); 2.5 U Taq polymerase; 0.45 ␮g (approximately 68 pmol) each 5⬘ and 3⬘ primers; and 1 ␮g DNA template. Cycle conditions were as follows: 40 1-minute cycles each at 93°C, 56°C, and 73°C followed by 1 7-minute extension cycle at 73°C for complete extension and annealing of PCR products. Seven ␮L PCR product was electrophoresed through 3% agarose TBE gel at 120 V for 2 hours. Bands were visualized by ethidium bromide staining. Expected product size was approximately 120 bp.

Radiolabeled Sequencing Gel Electrophoresis For greater band resolution, radiolabeled IgH PCR products were obtained by kinasing the upstream primer using T4 polynucleotide kinase (2 ␮L) in 10 ⫻ kinase buffer (2 ␮L) and 14,800 bq 33P ATP in a total reaction volume of 20 ␮L for 6 reactions. The radiolabeled PCR products were electrophoresed through 6% denaturing polyacrylamide gel at 60 W for 3 hours. The gel was vacuum-dried at 80°C and then autoradiographed overnight at ⫺70°C.

TCR Gene Rearrangements Amplification of TCR gene rearrangements was performed via nested PCR using sets of “outer” and “inner” V␥ and J␥ consensus primers corresponding to conserved V␥ and J␥ sequences.20 Each first-round reaction contained 2 ␮g DNA template; 50 mmol/L KCl; 10 mmol/L Tris-Cl at pH 8.0; 0.2 mmol/L each dNTPs; 0.01% gelatin; 1.5 mmol/L MgCl2; 40 pmol each “outer” primer; and 2.5 U Amplitaq DNA polymerase (Perkin-Elmer, Wellesley, MA). The PCR conditions were as follows: initial denaturation step of 2 minutes 30 seconds at 94°C; 25 1-minute cycles at 94°C; and an extension step of 1 minute 30 seconds at 55°C, 1 minute 30 seconds at 70°C, and 6 minutes at 70°C. Each second-round reaction contained 1/10 dilution of first-round products as template. Reactant and primer concentrations were the same as for the first round. The PCR cycle conditions for the second round were also the same except that only 15 cycles were performed. Five ␮L aliquot of each amplified product was electrophoresed through 1% TBE agarose gel to determine sufficient amplification and negative carryover. The sizes of the products for V␥1-8 and V␥9 TCR gene-rearranged products were approximately 420 bp and 380 bp, respectively. For denaturing gradient gel electrophoreses (DGGE) analysis, 50 ␮L of the final PCR products was precipitated in 3 mol/L sodium acetate and 80% ethanol and resuspended in 5 ␮L sterile water. The samples were denatured at 95°C for 5 minutes, then reannealed at 60°C for 1 hour before being loaded onto DGGE gel.20 Samples were electrophoresed at 60°C through 6.5% polyacrylamide gel containing a 30% to 60% urea/formamide denaturing gradient at 150 V. The gels were stained with ethidium bromide (1 mg/mL) for 15 min-

618

CUTANEOUS LYMPHOID HYPERPLASIA (Nihal et al)

utes, after which the gels were photographed under ultraviolet light.

Immunoperoxidase Methods Tissue specimens were snap-frozen in isopentane/dry ice, stored at ⫺70°C, cryostat sectioned, and stained with a 3-stage monoclonal antibody/biotin/avidin-horseradish peroxidase immunolabeling method as described previously.21 The monoclonal antibody panel included CD1a, CD2, CD3, CD4, CD5, CD7, CD8, CD19, CD20, CD22, CD23, CD25, CD43, and CD38. Controls included isotype-matched firststage antibodies of irrelevant specificity.

RESULTS Clinicopathologic Features Forty-four cases of CLH were analyzed during the study period. These included 38 typical cases with a mixed B-cell/T-cell composition and 6 cases containing ⱖ 90% T cells (T-cell cutaneous lymphoid hyperplasia [T-CLH]). Thirty-eight patients had solitary or localized lesions, including 4 cases of T-CLH, whereas 6 had regional or generalized lesions, including 2 cases of T-CLH. Forty cases were of idiopathic etiology. Suspected etiologies among the others included mercuric tattoo pigment, doxepin, clozapine, and bacterial infection. Patients were followed up for a mean and median period of 36.29 and 24 months, respectively (range, 6 to 104). During this time, there was no progression of disease except as described below for the 2 cases that evolved to overt CBCL. Analysis and classification of cases was based on an algorithm that incorporates histopathological, immunopathologic, and molecular biologic data (Figure 1). All cases contained a patchy to confluent dense dermal lymphoid infiltrate that generally spared the epidermis and was separated from it by a narrow Grenz zone. The infiltrate sometimes extended into the subcutis but usually had a top-heavy distribution with attenuation in the deeper portions of the specimen. In those cases composed predominantly of T cells (T-CLH), sometimes there was epidermal infiltration by lymphoid cells. B cells within the infiltrate were often organized into primary and secondary lymphoid follicles that were apparent as nodular aggregates at low power in paraffin sections. Otherwise, the infiltrate had a diffuse histological appearance. The cytological features of the infiltrate varied from case to case. There was generally a heterogeneous mixture of large and small lymphoid cells and histiocytes (macrophages, Langerhans cells, other dendritic cells). There were no confluent sheets of large lymphoid cells as seen in large-cell variants of CBCL. In those cases containing B-cell nodules that had formed secondary lymphoid follicles, these structures consisted of a germinal center containing a heterogeneous mixture of small cleaved and large noncleaved, mitotically active lymphoid cells, and tangible body macrophages surrounded by a mantle zone or cuff of small lymphocytes. The T cells in the interfollicular areas of the

FIGURE 1. Algorithm for the differential diagnosis of typical CLH mixed B-cell/T-cell type, clonal CLH, and CBCL. CLH, cutaneous lymphoid hyperplasia; CBCL, cutaneous B-cell lymphoma; IPOX, immunoperoxidase; IGH, immunoglobulin H; GR, gene rearrangement; PCR, polymerase chain reaction.

dermis were usually small and round, but they sometimes showed mild enlargement and mild nuclear irregularities. In many cases, plasma cells and eosinophils were also present. Immunoperoxidase analysis showed that all cases contained a mixture of B cells, T cells, macrophages, and dendritic cells. B cells and plasma cells were polytypic (expressed mixed ␬ and ␭ Ig light chains). They showed no abnormalities of B-cell surface antigen expression. The B cells were often organized into primary and secondary lymphoid follicles analogous to those occurring in reactive lymphoid tissues. These were frequently present regardless of whether they were recognized by routine histopathology. The follicular dendritic cell networks within B-cell follicles stained for both ␬ and ␭, consistent with the presence of polyclonal immune complexes on their cell surfaces. T cells were mixed CD4⫹⬎CD8⫹ without deficiency of any cell surface antigens except for occasional moderate deficiency of CD7 (20% to 40% positive). As a result of histopathological analysis, overt CBCLs of all types were excluded. As a result of immunopathologic analysis, cases of occult low-grade CBCLs (follicle center cell and marginal zone lymphomas) were also excluded. Molecular Biologic Analysis B-cell clonality was assessed using PCR assays to detect IgH gene rearrangements amplified with FR1/2

619

HUMAN PATHOLOGY

Volume 34, No. 6 (June 2003)

TABLE 1. Immunoglobulin and T-cell Receptor Gene Rearrangement in CLH Patients CLH Subtype CLH

T-CLH Total

Number of case

(Dual ⫹) IgH & TCR

Dominant clonal

Polyclonal

12 V␥(1-8)⫹ ⫽ 4 V␥(9)⫹ ⫽ 8 V␥(1-8)⫹V␥(9)⫹ ⫽ 3 1 V␥1-8⫹ ⫽ 1

2

23

15

0

4

2

13

2

27

17

IgH⫹

TCR⫹

38

9 FR1/2 ⫽ 7 FR3 ⫽ 2

6

3 FR1/2 ⫽ 2 FR3 ⫽ 1 12

44

and FR3 primers. T-cell clonality was determined using PCR/DGGE to detect TCR-␥ gene rearrangements amplified with V␥1-8 and V␥9 primers. Both sets of assays are approximately 80% sensitive. In addition, Southern blot analysis using a JH probe was performed in 5 cases. As summarized in Table 1, twenty-seven cases (61%) showed clonal CLH, including 12 IgH⫹ (27%; 3 cases of T-CLH); 13 TCR⫹ (30%; 1 case of T-CLH); and 2 IgH⫹/TCR⫹ (4%; neither case was T-CLH) (Figure 2). Of 12 IgH⫹ cases, 9 were positive by FR1/2 assay (Figure 3), and 3 were positive by FR3 assay (Figure 4). None of these cases was positive by both the assays. Among the TCR⫹ clonal CLH cases, 4 were V␥1-8⫹ alone; 8 were V␥9⫹ alone; and 3 were both V␥1-8⫹ and V␥9⫹. The remaining 17 cases were polyclonal (Figure 5). Five cases were analyzed using both conventional JH Southern blotting and PCR-based IgH assays. Two were polyclonal by both assays; 1 was clonal by both assays; 1 was clonal only by PCR; and 1 was clonal only by Southern blot. Among the 4 cases with suspected etiologies, 2 were IgH⫹ clonal. These included the case associated with clozapine (PCR⫹) and the mixed B-cell/T-cell CLH case associated with bac-

FIGURE 2. Diversity in the clonality patterns among cutaneous lymphoid hyperplasia cases analyzed for IgH gene rearrangements with FR1/2 and FR3 assays, TCR V␥1-8/V␥9 gene rearrangements, dual positivity, or a polyclonal pattern. IgH, immunoglobulin heavy chain; T-CLH, T-cell cutaneous lymphoid hyperplasia.

terial prostatitis (Southern blot ⫹), which resolved after administration of oral antibiotic therapy. Serial samples were available for molecular analysis in 7 cases, including the 2 lymphoma-associated cases discussed below. The remaining 5 cases showed consistent polyclonality in 2 cases; consistent IgH⫹ and TCR⫹ clonality in 1 case; change from IgH⫹ clonal to polyclonal in 1 case; and change from IgH⫹ clonal to polyclonal back to IgH⫹ clonal in 1 case. Two cases (4%; 1 case of T-CLH) progressed to CBCL. Both presented with regional distribution of lesions involving the scalp or back. The former (scalp) case progressed from polyclonal CLH by Southern blot analysis to follicle center cell type Ig⫺ IgH polyclonal CBCL during a 6-year period. During this time, the lesions became less sensitive to intermittent intralesional corticosteroids. At one point, a reactive 1-cm retroauricular lymph node was removed. The CBCL regressed completely after administration of local radiation therapy. The latter (back) case progressed from IgH⫹ clonal T-CLH to IgH⫹, Ig⫺ follicle center cell type CBCL during a 1-year period. During this time, lesions became unresponsive to intralesional corticosteroids and progressively enlarged to form multiple large tumors. There was a matching IgH clonal band pattern between the 2 specimens as shown by the IgH FR1/2 assay using radiolabeled primers (Figure 6). There was a complete clinical response to local radiation therapy.

FIGURE 3. Agarose gel showing IgH polymerase chain reaction gene products with the FR1/2 assay. M, marker lane; 1 and 2, representative patient lanes showing (⫹) or (⫺) band patterns; P, positive control cell line MC/CAR; T, tonsil for polyclonal control; CO, reagent negative control (no DNA).

620

CUTANEOUS LYMPHOID HYPERPLASIA (Nihal et al)

FIGURE 4. Agarose gel showing IgH gene polymerase chain reaction products with the FR3 assay. M, the marker lane; 1, 2, and 3, representative patient lanes showing (⫹) or (⫺) band patterns; IgH, immunoglobulin heavy chain; P, positive control cell line MC/CAR; T, tonsil polyclonal control; CO, reagent negative control (no DNA).

DISCUSSION In the current series of CLH cases, dominant clonality was detected in 27 of 44 (61%) using PCRbased assays for IgH and TCR gene rearrangements (Table 1). In general, dominant clonality can be deciphered by 1 or 2 discrete and distinct bands in a single lane that represents the rearranged alleles of IgH or TCR␥ genes within a dominant lymphoid clonal population. We have referred to such cases as “clonal CLH.”2,15This prevalence of clonal CLH is higher than noted in previous studies: however, most of these detected clonality using Southern blotting15,16 which is less sensitive than PCR-based methods.22,23 PCR-based assays also vary in their sensitivity (detected positives relative to true positives) and their clonal detection thresholds (the minimum percentage of clonal cells detectable). Our PCR/DGGE assay is approximately 80% sensitive, with a clonal detection threshold of approximately 1%.24,25 Our IgH PCR assays have a combined sensitivity of approximately 80% and a clonal detection threshold of approximately 10% when abundant B cells are present.26 From a technical standpoint, it is important to emphasize the potential for misdiagnosis of dominant clonality in infiltrates containing only sparse B cells. The detection threshold is enhanced to ⬍ 1% when only sparse B cells are present because the rare B cells in such samples can be randomly amplified in

FIGURE 5. Denaturing gradient gel electrophoreses gel showing polyclonal banding patterns for T-cell receptor V␥1-8 gene polymerase chain reaction products in patient lanes (1 through 4) showing polyclonal smear (⫺). P, positive control cell line Jurkat; T, tonsil polyclonal control; CO, reagent negative control (no DNA).

FIGURE 6. Six percent polyacrylamide denaturing sequencing gel showing a clonal band pattern within the polyclonal background, using radiolabeled primers, in serial biopsy specimens of the same patient during 1-year period. Lanes 1 through 3 and 4 through 6 are triplicate runs of individual sample lanes, followed by their mixes (Mix). P, positive control cell line Daudi; T, tonsil polyclonal control; CO, reagent negative control (no DNA).

early cycles of the PCR, thus resulting in false-positive clonal bands. This can be seen in normal skin specimens and infiltrates containing only sparse B cells.26 We recently showed that triplicate analysis of samples will control for this artifact because true-positive bands will be consistent in size, whereas false-positive bands will not (Figure 6).27 Although theoretically possible for T cells as well, this artifact is uncommon using PCR/ DGGE, probably because of the relative abundance of T cells in most skin biopsy specimens. We cannot determine the origin of each of the dominant gene rearrangements detected in our cases. Among those clonal CLH cases with either IgH or TCR but not both, it is likely that dominant IgH gene rearrangements are caused by a B-cell clone, and dominant TCR gene rearrangements are caused by a T-cell clone. Among those clonal CLH cases with both IgH and TCR gene rearrangements, it is possible that the infiltrates harbor more than 1 dominant clone, one of which might be driving or supporting the growth of the other. Alternatively, a single clone might contain both IgH and TCR gene rearrangements as has been observed in other lymphomas.28 Two of our 44 patients with CLH (4%) developed overt CBCL within a few years of onset. One of these was clonal CLH and contained the same dominant IgH gene rearrangement in the CLH and subsequent CBCL lesions. The lack of detectable clonality in the other case was probably related to the inability of the PCR methods to detect all cases harboring dominant clonality. The IgH and TCR assays we employed have a falsenegative rate of approximately 20% each.24,26 This view was also supported by the continued lack of detectable clonal IgH gene rearrangement in the overt CBCL that developed subsequently in this patient. The existence of a monoclonal prelymphomatous process, such as clonal CLH, is not unique to CLH; it also occurs in the context of lymphoid hyperplasia in other tissues, lymphomatoid papulosis, angioimmunoblastic lymphadenopathy with dysproteinemia, and the so-called benign monoclonal gammopathies (now generally referred to as “monoclonal gammopathies of uncertain significance”).20 In this context, it is interesting to note that clonal CLH was detected among 2 of the 4 CLH cases with a probable known antigenic trigger

621

HUMAN PATHOLOGY

Volume 34, No. 6 (June 2003)

based on clinical history: one associated with clozapine and one associated with bacterial prostatitis, which resolved with antibiotic therapy. These observations suggest that clonal B-cell proliferations in the skin may arise from exogenous antigenic triggers. Parallel situations exist for gastric mucosa-associated lymphoid tissue lymphomas associated with Helicobacter pylori, posttransplant lymphoproliferative disorders associated with Epstein-Barr virus, and some cases of European CBCL associated with Borrelia burgdorferi.29,30 Each of these disorders resolved with clearance of the inciting agent either with antibiotic therapy or reversal of immunosuppression.10 Other studies have shown that, occasionally, certain atypical cutaneous lymphoid infiltrates secondary to drugs may also contain dominant T-cell clones.11 In aggregate, these observations suggest that infectious agents and drugs can be direct causes of clonal CLH that may progress to overt lymphoma and yet remain antigen driven and potentially curable. We and others3,15 have observed the progression of clonal CLH to overt B-cell or T-cell lymphoma. Although many cases of CLH are undoubtedly true hyperplasias, our findings indicate that in many other cases, CLH is not really a hyperplasia at all but rather a clonal lymphoproliferative disorder that may progress occasionally to overt cutaneous lymphoma. These observations help explain the pathogenesis of CLH and its known association with cutaneous lymphomas. Although the detection of dominant clonality did not predict a greater likelihood of developing lymphoma in our series (1 of 27 clonal versus 2 of 44 overall), with greater case numbers, more sensitive clonality assays, and longer follow-up, it may prove to be a relevant risk factor. REFERENCES 1. Ploysangam T, Breneman DL, Mutasim DF: Cutaneous pseudolymphomas. J Am Acad Dermatol 38:877-895, 1998 2. Wood GS: Cutaneous lymphoid hyperplasia. In Arndt KE, LeBoit PE, Robinson JK, Wintroub BU, (eds): Cutaneous Medicine and Surgery: An Integrated Program in Dermatology, ed 1. Philadelphia: Saunders, 1996, pp 1626-1630 3. Wood GS: Inflammatory diseases that stimulate lymphomas: Cutaneous pseudolymphomas. In Freedberg IM, Eisen AZ, Wolff K, et al: Dermatology in General Medicine, ed 5. New York, McGrawHill, 1998, pp 1259-1274 4. Allen AC: Persistent “insect bites” (dermal eosinophilic granulomas) simulating lymphoblastomas, histiocytoses and squamous cell carcinomas. Am J Pathol 24:367-387, 1948 5. Allen AC: Reactions to arthropodes. In: The Skin, ed 2. New York: Grune and Stratton, 1967, pp 561-568 6. Barr-Nea L, Sandbank M, Ishay J: Pseudolymphoma of skin induced by oriental hornet (Vespa orientalis) venom. Experientia 32:1564-1565, 1976 7. Barr-Nea L, Ishay J: Histopathological changes in mouse and rat skin injected with venom sac extracts of the oriental hornet (Vespa orientalis). Toxicon 15:301-306, 1977 8. Sandbank M, Barr-Nea L, Ishay J: Pseudolymphoma of skin induced by oriental hornet (Vespa orientalis) venom ultrastructure study. Arch Dermatol Res 262:135-141, 1978 9. Brady SP, Magro CM, Diaz-Cano SJ, et al: Analysis of clonality of atypical cutaneous lymphoid infiltrates associated with drug therapy by PCR/DGGE. Hum Pathol 30:130-136, 1999

10. Magro CM, Crowson AN: Drugs with antihistaminic properties as a cause of atypical cutaneous lymphoid hyperplasia. J Am Acad Dermatol 32:419-428, 1995 11. Magro CM, Crowson AN, Schapiro BL: The interstitial granulomatous drug reaction: A distinctive clinical and pathological entity. J Cutan Pathol 25:72-78, 1998 12. Picken RN, Strle F, Ruzic-Sabljic E, et al: Molecular subtyping of Borrelia burgdorferi sensu lato isolates from five patients with solitary lymphocytoma. J Invest Dermatol 108:92-97, 1997 13. Garcia CF, Weiss LM, Warnke RA, et al: Cutaneous follicular lymphoma. Am J Surg Pathol 10:454-463, 1986 14. Medeiros LJ, Picker LJ, Abel EA, et al: Cutaneous lymphoid hyperplasia: Immunologic characteristics and assessment of criteria recently proposed as diagnostic of malignant lymphoma. J Am Acad Dermatol 21:929-942, 1989 15. Wood GS, Ngan BY, Tung R, et al: Clonal rearrangements of immunoglobulin genes and progression to B-cell lymphoma in cutaneous lymphoid hyperplasia. Am J Pathol 135:13-19, 1989 16. Hammer E, Sangueza O, Suwanjindar P, et al: Immunophenotypic and genotypic analysis in cutaneous lymphoid hyperplasia. J Am Acad Dermatol 28:426-433, 1993 17. Myers RM, Maniatis T, Lerman LS: Detection and localization of single base changes by denaturing gel electrophoresis. Methods Enzymol 155:501-527, 1987 18. Hummel M, Ziemann K, Lammert H, et al: Hodgkin’s disease with monoclonal and polyclonal populations of Reed-Sternberg cells. N Engl J Med 333:901-906, 1995 19. Slack DN, McCarthy KP, Wiedemann LM, et al: Evaluation of sensitivity, specificity, and reproducibility of an optimized method for detecting clonal rearrangements of immunoglobulin and T-cell receptor genes in formalin-fixed, paraffin-embedded sections. Diagn Mol Path 2:223-232, 1993 20. Wood GS, Tung RM, Haeffner AC, et al: Detection of clonal T-cell receptor ␥ gene rearrangements in early mycosis fungoides/ sezary syndrome by polymerase chain reaction and denaturing gradient gel electrophoresis (PCR/DGGE). J Invest Dermatol 103:34-41, 1994 21. Wood GS, Warnke R: The immunophenotyping of bone marrow biopsies and aspirates: Frozen section techniques. Blood 59:913-922, 1982 22. Bouloc A, Delfau-Laure MH, Lenormand B, et al: Polymerase chain reaction analysis of immunoglobulin gene rearrangement in cutaneous lymphoid hyperplasias. Arch Dermatol 135:168-172, 1999 23. Ceballos KM, Gascoyne RD, Martinka M, et al: Heavy multinodular cutaneous lymphoid infiltrates: Clinicopathologic features and B-cell clonality. J Cutan Pathol 29:159-167, 2002 24. Mikkola D, Horvath N, Gilliam AC, et al: Consistently high sensitivity of PCR/DGGE analysis for the detection of dominant clonality in cutaneous T-cell lymphomas (CTCLs). J Invest Dermatol 112:627, 1999 25. Wood GS, Uluer AZ: Polymerase chain reaction/denaturing gradient gel electrophoresis (PCR/DGGE): Sensitivity, band pattern analysis and methodological optimization. Am J Dermatopathol 21: 547-551, 1999 26. Nihal M, Mikkola D, Wood GS: Detection of clonally restricted immunoglobulin heavy chain gene rearrangements in normal and lesional skin: Analysis of the B-cell component of the skin-associated lymphoid tissue and B-cell lymphomas. J Mol Diagn 2:5-10, 2000 27. Nihal M, Mikkola DL, Qian Z, et al: The clonality of tumorinfiltrating lymphocytes in African Kaposi’s sarcoma. J Cutan Pathol 28:200-205, 2001 28. Knowles DM, Chadburn A: Lymphadenopathy and the lymphoid neoplasms associated with the acquired immunodeficiency syndrome. In: Neoplastic Hematopathology, ed 2. Philadelphia: William and Wilkins, 2001, pp 682-684, 971-975, 1052 29. Cerroni L, Zochling N, Putz B, et al: Infection by Borrelia burgdorferi and cutaneous B-cell lymphoma. J Cutan Pathol 24:457461, 1997 30. Garbe C, Stein H, Dienemann D, et al: Borrelia burgdorferi associated cutaneous B cell lymphoma: Clinical and immunohistochemical characterization of four cases. J Am Acad Dermatol 24:584590, 1991

622