Primary cutaneous B-cell lymphoma

Primary cutaneous B-cell lymphoma

CLINICAL REVIEW Primary cutaneous B-cell lymphoma Melissa A. Bogle, MD, Christy C. Riddle, MD, Emily M. Triana, MD, Dan Jones, MD, PhD, and Madelein...

257KB Sizes 0 Downloads 14 Views

CLINICAL

REVIEW

Primary cutaneous B-cell lymphoma Melissa A. Bogle, MD, Christy C. Riddle, MD, Emily M. Triana, MD, Dan Jones, MD, PhD, and Madeleine Duvic, MD Houston, Texas Primary cutaneous B-cell lymphomas include extranodal marginal zone B-cell lymphoma, follicular lymphoma, large B-cell lymphoma, and, rarely, mantle cell lymphoma. Our purpose in conducting this review was to determine the clinical and behavioral characteristics of primary cutaneous B-cell lymphomas, their relationship to infectious triggers, and therapeutic response. We conducted a retrospective chart review of 23 adult patients presenting to the dermatology clinic at M. D. Anderson Cancer Center with primary cutaneous B-cell lymphoma between January 1999 and May 2003. Primary cutaneous B-cell lymphomas generally present on the head and neck, with the trunk and extremities afflicted to a lesser extent. Patients were found to have serologic evidence of prior infection with Borrelia burgdorferi (n = 10), Helicobacter pylori (n = 5), and Epstein-Barr virus (n = 6). Overall, treatment of primary cutaneous B-cell lymphoma should involve multiple modalities; however, specific treatment aimed at concurrent or suspected infection, particularly B burgdorferi, is a helpful adjunct and may achieve complete remission in a small subset of patients. ( J Am Acad Dermatol 2005;53:479-84.)

C

utaneous B-cell lymphomas can present as primary cutaneous tumors or as secondary to systemic lymphomas. Classification of primary cutaneous B-cell lymphomas is controversial and can be organized according to the World Health Organization (WHO), Revised European-American Lymphoma, or European Organization for Research and Treatment of Cancer (EORTC) classification systems.1 We utilize the WHO classification, which divides primary cutaneous B-cell lymphomas into follicular lymphoma, large B-cell lymphoma, extranodal marginal zone B-cell (MALT) lymphoma, and mantle cell lymphoma.2 Each type represents distinct clinical and histopathological subtypes of extranodal lymphoma, although overlap does occur.

METHODS The M. D. Anderson Cancer Center dermatology clinic database was queried to identify patients older than 16 years of age who had been given the diagnosis of primary cutaneous B-cell lymphoma

From the Departments of Dermatology and Hematopathology, University of Texas and M. D. Anderson Cancer Center. Funding sources: None. Conflicts of interest: None identified. Reprint requests: Madeleine Duvic, MD, Department of Dermatology, M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 0434, Houston, TX 77030. E-mail: [email protected]; [email protected] hotmail.com. 0190-9622/$30.00 ª 2005 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2005.04.043

Abbreviations used: CHOP:

cyclophosphamide, doxorubicin, vincristine, and prednisolone CR: complete response EORTC: European Organization for the Research and Treatment of Cancer LTFU: lost to follow-up MALT: marginal zone B-cell (lymphoma) PCR: polymerase chain reaction PR: partial response WHO: World Health Organization

from January 1999 to May 2003. All patients underwent a staging work-up for systemic lymphoma, including complete blood cell count, chemistry panel, b2-microglobulin, computed tomographic scans, bone marrow aspirate and/or trephine biopsy with gene rearrangement studies, and serologic infectious disease panel. Secondary cutaneous Bcell lymphomas were excluded from this study; all presented with extracutaneous involvement at their initial systemic work-up. Chart reviews were performed under Institutional Review Board approval to determine demographic information as well as associated medical conditions, known infections, therapy, and response. There was uniform review of the skin biopsies to confirm the diagnosis by one of the coauthors (D. J.) using the WHO classification scheme.3 Paraffinsection immunohistochemistry was done on all cases for at least one B-cell marker, usually CD20, and at least one T-cell marker, usually CD3, CD5, or both. In cases with a follicular growth pattern, 479

480 Bogle et al

J AM ACAD DERMATOL SEPTEMBER 2005

Fig 1. Clinical manifestations of cutaneous B-cell lymphoma. A, Marginal zone lymphoma of the scalp with purple nodules and plaques. B, Marginal zone lymphoma with follicular accentuation presenting on the back. C, Large B-cell lymphoma of the lower leg. D, Large B-cell lymphoma presenting as pink papule on neck.

Bcl-2 immunostaining was done in all cases, along with CD10 or Bcl-6 in some cases. Immunoglobulin k and l light chain staining was done in cases with a monocytoid/plasmacytoid lymphoid component. In the follicular lymphomas and in the marginal zone lymphomas, immunoglobulin heavy chain gene rearrangement studies were performed by polymerase chain reaction (PCR) in 9 cases using framework-1, -2 and -3 VH-J region primer sets, as previously described.4 Quantitative real-time PCR analysis for JH/BCL2 fusion products involving the major breakpoint region (mbr) or minor cluster region (mcr) was done in 12 cases, as previously described.5 On the basis of these studies and the histologic features, extranodal marginal zone B-cell (MALT) lymphoma was diagnosed in cases with a predominance of small CD201 B cells and variable numbers

of monocytoid and plasmacytoid forms. Follicular lymphoma was diagnosed unequivocally in cases with JH/BCL2 fusion products or uniformly bcl2epositive lymphocytes in follicles or in those bcl-2enegative tumors that were overwhelmingly composed of well-formed follicles that lacked features of reactive follicles. Overall, all but one lymphoma showed evidence of monotypic light chain and/or clonal IgH PCR rearrangements and/or IgH/BCL2 fusion products. The remaining case, which was negative by IgH PCR, had limited material for further analysis but was diagnostic of lymphoma based on histologic features.

RESULTS Between January 1999 and May 2003, 23 patients presented to the Dermatology Clinic with primary

Bogle et al 481

J AM ACAD DERMATOL VOLUME 53, NUMBER 3

Fig 2. Number of primary cutaneous B-cell lymphoma lesions at each anatomic location. *Numbers are greater than total number of patients because some patients had tumors in more than one location.

cutaneous B-cell lymphomas. Tumors were classified according to the WHO classification as marginal zone lymphoma (n = 11), follicular lymphoma (n = 7), and large B-cell lymphoma (n = 5). No patients in this series had mantle cell lymphoma. The male (n = 12) to female (n = 11) ratio was 1:1. The mean patient age was 53.8 years (range, 24-83 years). The majority of patients were white (n = 17); 1 was black and 5 were of Hispanic origin. Lesions of primary cutaneous B-cell lymphoma generally presented as pink to violaceous papules, plaques, and nodules, ranging from 0.2 to 9 cm (Fig 1). Three patients with follicular cell lymphoma presented with subtle purple to pink patches on the scalp or temples, whereas the remaining patients presented with nodules or small translucent papules several millimeters in diameter on the arms (n = 2), forehead or chin. Two patients with large B-cell lymphoma presented with shiny purple-red plaques on the leg, whereas 3 presented with violaceous nodules on the nose, back, or leg. Ulceration was common in lesions on the scalp and extremities. Six of the marginal zone lymphomas presented as pink to red plaques, papules, or nodules, one with ‘‘gray follicular dots’’ visible within the plaque. These tumors presented on the cheek/temple, neck, axillary fold, chest, back, and, in one unusual case, all 4 extremities concurrently. The other marginal zone lymphomas were violaceous plaques or nodules located on the scalp, cheek, arm (n = 2), or back. As shown in Fig 2, the most common location for primary cutaneous B-cell lymphomas overall was the

Table I. Patients with cutaneous B-cell lymphoma who were seropositive for various infections Seropositivity* Lymphoma subtype

Follicular lymphoma Large B-cell lymphoma Marginal zone lymphoma

Total No. of patients tested

B burgdorferi

H pylori

EBV

CMV

4

1

2

2

1

3

1

0

1

1

11

8

3

3

2

CMV, Cytomegalovirus; EBV, Epstein-Barr virus. *Number of patients who were seropositive for each infection.

head and neck (n = 11), including 4 marginal zone lymphomas, 2 large B-cell lymphomas, and 5 follicular lymphomas. Marginal zone lymphomas were the most common variant on the trunk in this series (3/5 cases). Two patients had large B-cell lymphomas on the trunk (n = 1) or head and neck (n = 1). Large B-cell lymphoma involving the leg, a cutaneous lymphoma considered to be a separate entity by the EORTC, was present in 3 patients. Immunohistochemical studies revealed that 7 of the cases displayed significant numbers of admixed CD31 reactive T cells, including one follicular lymphoma, one large B-cell lymphoma, and 5 marginal zone lymphomas. Seven of the 13 tested skin biopsy specimens had demonstrable monotypic light chain staining patterns, including k light chain in one follicular lymphoma and in 5 marginal zone

482 Bogle et al

J AM ACAD DERMATOL SEPTEMBER 2005

Table II. Therapeutic response of primary cutaneous B-cell lymphomas Antibiotic 1 Antibiotic Local IL IL Antibiotic alone XRT triamcinolone triamcinolone 1 XRT

Follicular LTFU (3) CR lymphoma Large B-cell lymphoma CRz LTFU Marginal zone LTFU lymphoma NCRz

CR

Permethrin/ Rituximab, Rituximab, sulfacetamide CHOP CHOP IFN FND 1 CHOP 1 XRT 1 XRT alfa 1 XRT imiquimod cream

PRy

CR* CR (2)

CR

CR

CR

PR

PR

CR

Three patients are not included in the treatment table. One patient with large B-cell lymphoma of the leg refused all treatment; two patients with marginal zone lymphoma have been recently diagnosed and treatment options are still under consideration. CHOP, Cyclophosphamide, doxorubicin, vincristine, and prednisolone; CR, complete response; FND, fludarabine, mitoxantrone (Novantrone), dexamethasone; IFN, interferon; IL, intralesional; LTFU, lost to follow-up; NCR, near clear response; PR, partial response; XRT, radiation therapy. *H pylori seropositivity. y Patient had biopsy-proven pseudolymphoma lesions on and off for the past 14 years that were thought to be driven by chronic Demodex folliculitis before her diagnosis of follicular cell lymphoma. z B burgdorferi, H pylori seropositivity.

lymphomas, and l light chain staining in one marginal zone lymphoma. The remaining cases had indeterminate staining. Six of 7 patients with primary cutaneous follicular lymphoma showed bcl-2 expression in follicular B cells, and 3 of 7 of them had JH/BCL2 fusion products identified by PCR.6-8 Clonal IgH rearrangements were detected by PCR in 7 of 9 tested patients with marginal zone or follicular lymphoma. One of the two cases lacking detectable IgH rearrangements had an identifiable JH/BCL2 fusion. Table I lists serological results of past exposure to lymphoma-associated organisms by tumor type. Borrelia burgdorferi seropositivity was found in 8 of 11 patients with marginal zone lymphoma, 1 of 4 patients with follicular lymphoma, and 1 of 3 patients with large B-cell lymphoma. Helicobacter pylori infection was found in 3 of 11 patients with marginal zone lymphoma and in 2 of 4 patients with follicular lymphoma. We also checked for exposure to common DNA viruses (Table I). None of the patients with follicular lymphoma had developed nodal metastases in the average follow-up period of 13 months (range, 1-41 months). None of the large B-cell lymphoma had extracutaneous involvement at a follow-up of 14 months as with the marginal zone lymphomas, which had an average of 29 months’ follow-up (range, 12-55 months). As shown in Table II, local conservative therapies were given for indolent B-cell lymphomas in 12 patients including antibiotics for seropositive patients (n = 6; 1 complete response [CR], 1 partial response [PR], 4 lost to follow-up [LTFU]), intrale-

sional corticosteroids (1 CR for 3 years), antibiotics plus steroids (1 CR), imiquimod (1 PR), or the more traditional therapy of local radiation to tumor field (n = 2; 1 CR, 1 LTFU) plus antibiotics (1 CR). One patient with marginal zone lymphoma had human papillomavirus and chronic active Epstein-Barr virus seropositivity (IgG 7.17 g/L, IgM 2.5 g/L) and had a PR to interferon alfa given 3 million U 3 times weekly. Six patients received chemotherapy. Regimens included cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP) (2 CRs), CHOP with radiation (2 CRs), or triple therapy with CHOP, rituximab, and radiation (1 PR, in a patient with large B-cell lymphoma). One patient with marginal zone lymphoma was treated with the use of fludarabine, mitoxantrone (Novantrone), and dexamethasone with rituximab plus radiation and achieved CR.

DISCUSSION Demographic data in our series revealed an equal incidence of primary cutaneous B-cell lymphoma among male and female patients. This finding differs from those of the EORTC, which found a female/ male ratio of 2:1.9,10 The average age at diagnosis in our population was similar to that found in other reports of onset in the late fifties.9,11 The range of patient ages was comparable to published results of 22 to 92 years.12 We found a bimodal distribution of age at presentation, with the majority of patients in their thirties or seventies. As has been reported, tumors generally presented on the head and neck with the trunk and extremities afflicted to a lesser extent.10,11,13

J AM ACAD DERMATOL

Bogle et al 483

VOLUME 53, NUMBER 3

Chromosomal abnormalities play a significant role in some nodal B-cell lymphomas. The (14;18)(q32; q21) translocation is found in 70% to 90% of follicular nodal lymphomas and has rarely been reported in cutaneous follicular lymphoma.9 Bcl-2 is the gene product of this translocation. Studies have shown bcl-2 expression in primary and secondary cutaneous lymphomas, but the value of this determination is still under investigation.9,14 In this study, bcl-2 expression was found most consistently in primary cutaneous follicular cell lymphoma, appearing in 6 of 7 tumors. To our knowledge none of the patients ever went on to have extracutaneous involvement. Various infections have been associated with the development of primary cutaneous B-cell lymphomas. Lyme disease has been linked to cutaneous B-cell lymphoma with indolent behavior, namely immunocytomas, now considered marginal zone lymphomas of the MALT type.2,13 A previous study by Cerroni et al15 found 18% of paraffin-embedded tissue sections of primary cutaneous B-cell lymphoma to harbor B burgdorferi DNA. These were identified in various types of B-cell lymphomas, including follicle center lymphoma (n = 3/20), immunocytoma (n = 3/4), marginal zone lymphoma (n = 2/20), and diffuse large B-cell lymphoma (n = 1/6).15 In this small US study, seropositivity for previous exposure to B burgdorferi was most closely associated with marginal zone lymphoma and was present in 8 of 11 (73%) of all marginal zone lymphoma cases. Persons living in areas endemic for Lyme disease in the United States have a seroprevalence of antibodies to B burgdorferi of approximately 10.9%.16 H pylori infection is often found in MALT lymphomas of the gastric mucosa and intestine, and remission has been induced by treatment of the underlying infection.17 This may also hold true for a subset of patients with cutaneous B-cell lymphomas. We found H pylori seropositivity in 3 of 11 patients with marginal zone lymphoma (who also were positive for B burgdorferi) and 2 of 4 patients with follicular lymphoma. The prevalence of antibodies to H pylori in the US general population is approximately 10% in persons younger than 35 years of age, with rates increasing to almost 80% by age 75.18 Two patients in this study, both with marginal zone lymphoma, achieved CR or near CR to antibiotic therapy alone. The first patient is a 35-year-old Hispanic woman who presented with pink papules on the shoulders and back. She was found to have seropositivity for H pylori and previous exposure to B burgdorferi. Her lesions resolved completely after antieH pylori treatment with Prevpac (oral

lansoprazole 30 mg twice daily, oral clarithromycin 500 mg twice daily, and oral amoxicillin 1 g twice daily for 2 weeks) combined with oral tetracycline 500 mg twice daily for 2 months. The patient remains clear at 3 years’ follow-up with occasional periods of urticaria that has been responsive to additional Prevpacs. The second patient is a 34-year-old Hispanic man who contracted Lyme disease after being bitten by a tick at a California state park. At that time his Borrelia titer was 1:512, and he underwent a full course of antibiotic therapy prescribed by his physician in Mexico. Three years later pink annular rings with infiltrated papules developed in exactly the same sites as his erythema chronicum migrans. Biopsy findings of the lesions revealed a marginal zone lymphoma. The patient achieved a near CR with oral doxycycline 100 mg twice daily for 6 months, then daily to 1 year. At 1 year his lesions began to turn pink again, and the antibiotic was switched to oral ampicillin 500 mg twice daily, again with a near CR. The patient remains on a regimen of antibiotic therapy. In conclusion, overall treatment of primary cutaneous B-cell lymphoma may involve multiple modalities; however, there appears to be a subset of patients who may achieve complete remission with treatment aimed at concurrent or suspected infection that may be driving their disease. A few patients (2/23 in our series, or 9%) will achieve clearing with antibiotic therapy alone. Staging work-up should routinely include infectious disease screening because a few patients will also derive benefit from adjunctive antibiotic therapy for contributing infections. REFERENCES 1. Prince HM, O’Keefe R, McCormack C, Ryan G, Turner H, Waring P, Baker C. Cutaneous lymphomas: which pathological classification? Pathology 2002;34:36-45. 2. Cerroni L, Helmut K, Gatter K. An illustrated guide to skin lymphoma. Oxford (UK): Blackwell Science; 1998. 3. Jaffe ES, Harris NL, Stein H, Vardiman JW. Tumors of haematopoietic and lymphoid tissue. Lyon, France: IARC Press; 2001. 4. Abruzzo LV, Rosales CM, Medeiros LJ, Manning JT, Keating MJ, Jones D. Transient Epstein-Barr virus-positive large B-cell lymphoproliferative disorders arising in immunodeficient patients previously treated with fludarabine for low-grade B-cell lymphoma. Am J Surg Pathol 2002;26:630-6. 5. Bowman A, Jones D, Medeiros LJ, Luthra R. Quantitative detection of t(14;18) bcl-2/JH fusion sequences in follicular lymphoma patients: comparison of peripheral blood and bone marrow aspirate samples. J Mol Diagn 2004;6:396-400. 6. Cerroni L, Volkenandt M, Rieger E, Soyer HP, Kerl H. Bcl-2 protein expression and correlation with the interchromosomal 14:18 translocation in cutaneous lymphomas and pseudolymphomas. J Invest Dermatol 1994;102:231-5. 7. Hammer E, Sangueza O, Suwanjindar P, White CR Jr, Braziel RM. Immunophenotypic and genotypic analysis in

484 Bogle et al

J AM ACAD DERMATOL SEPTEMBER 2005

8.

9.

10.

11.

12.

cutaneous lymphoid hyperplasias. J Am Acad Dermatol 1993;28:426-33. Wood GS, Ngan BY, Tung R, Hoffman TE, Abel EA, Hoppe RT, et al. Clonal rearrangements of immunoglobulin genes and progression to B-cell lymphoma in cutaneous lymphoid hyperplasia. Am J Pathol 1989;135:13-9. Pandolfino T, Siegel R, Kuzel T, Rosen S, Guitart J. Primary cutaneous B-cell lymphoma: review and current concepts. J Clin Oncol 2000;18:2152-68. Burg G, Kempf W, Haeffner A, Nestle F, Hess-Schmid M, Doebbeling U, et al. Cutaneous lymphomas. Curr Probl Dermatol 1997;9:137-204. Sarris A, Braunschweig I, Medeiros L, Duvic M, Ha CS, Rodriguez MA, et al. Primary cutaneous non-Hodgkin’s lymphoma of Ann Arbor Stage I: preferential cutaneous relapses but high cure rate with doxorubicin-based therapy. J Clin Oncol 2001;19:398-405. Pimpinelli N, Santucci M, Carli P, Paglierani M, Bosi A, Moretti S, et al. Primary cutaneous follicular lymphoma:

13. 14.

15.

16.

17.

18.

clinical and histological aspects. Curr Probl Dermatol 1990; 19:203-20. Willemze R. Primary cutaneous lymphomas. Curr Opin Oncol 2000;12:419-25. Triscott JA, Ritter JH, Swanson PE, Wick MR. Immunoreactivity for bcl-2 protein in cutaneous lymphomas and lymphoid hyperplasias. J Cutan Pathol 1995;22:2-10. Cerroni L, Zochling N, Putz B, Kerl H. Infection by Borrelia burgdorferi and cutaneous B-cell lymphoma. J Cutan Pathol 1997;24:457-61. Jochimsen E, Sikkink J, Marx JJ. The prevalence of Borrelia burgdorferi seropositivity in an area endemic for Lyme disease. Wis Med J 1990;89:677-81. Arima N, Tsudo M. Extragastric mucosa-associated lymphoid tissue lymphoma showing the regression by Helicobacter pylori eradication therapy. Br J Haematol 2003;120:790-2. Marshall BJ. Epidemiology of H. pylori in Western countries. In: Hunt RH, Tytgat GNJ, editors. Helicobacter pylori: basic mechanisms to clinical cure. Boston: Kluwer Academic; 1994. pp. 75-84.