Case 1 -

Lymphomatoid Drug Eruption A. Neil Crowson Regional Medical Lab Tulsa, OK

Click on each slide thumbnail image for an enlarged view

Clinical History A 74 year old woman developed a psoriasiform, papular and plaque-like eruption of the back, arms, legs, palms and soles whilst on Cozaar (losarten – a potent angiotensin-type I receptor blocker), atenolol (a beta-blocker), the diuretic hydrochlorothiazide and verapamil (a calcium channel blocker). There was no involvement of facial skin and no lymphadenopathy was detectable. Apart from severe arterial hypertension, an examination and review by an internist showed no systemic disease. An initial skin biopsy in December 2000 showed a superficially disposed atypical small lymphoid infiltrate with coarse sclerosis of the collagen table, in concert with an epidermotropic atypical lymphoid infiltrate suggesting large plaque parapsoriasis in evolution to mycosis fungoides. Flow cytometry of peripheral blood showed changes consistent with a reactive process by virtue of a Th/s ration of 2,6:1 with normal expression of CD2,3,5 and7 with no aberrant loss of surface antigen expression; no Sezary cells were detectable. After subsequent enquiries concerning the drug history, the patient was referred to a dermatologist with a specific interest and training in pharmacotherapeutics, and a second biopsy was performed in January 2001 following by modulation of the drug regime in concert with topical steroid therapy (0.1% triamcinolone). As the last agent added to the patient's drug was Cozaar, that drug was discontinued first, in concert with the thiazide. The second biopsy pair showed similar features to the first, but with important distinctions. First, the intensity of the infiltrate was much reduced, as was the degree of epidermotropism. Secondly, the involvement of the epidermis was patchy as opposed to continuous. At follow up 2 months later the rash was slightly improved and she commenced tanning bed therapy for 3 minutes 3-4 times week. At follow up 1 month later new lesions were still developing and the atenolol was then discontinued. Three weeks later the eruption was markedly improved with a reduction of the involved surface area by 80%. She then restarted Cozaar under the guidance of her internist, with an immediate and dramatic flare of the eruption on examination in June 2001. The Cozaar was again discontinued and by July the rash was largely resolved. In September the verapamil was also discontinued. By October 2001 all that remained of the rash were two small eczematous patches on the back. Clonidine was substituted for the prior anti-hypertensive therapy and atenolol was re-introduced. The tanning bed was tapered. The conclusion was that cozaar and verapamil were synergistic in provocation of the eruption. Her hypertension was unstable and 18 months later an alternate calcium channel blocker and beta blocker (Norvasc and Toprol respectively) were substituted with flaring of the eruption. A rebiopsy at that time (January 2003) showed changes held to be compatible with a lymphomatoid drug eruption. In particular, there was directed migration of small lymphoid forms into areas of maximum antigenic processing, namely, acrosyringia and hair follicles, with a significant reduction in the intensity of the infiltrate. The calcium channel blocker was discontinued, methyldopa was substituted, and the rash resolved. She remains in complete dermatologic remission, with no evidence of the rash on the previously involved trunk at the last examination in October of 2004. Case 1 - Figure 1A - The punch biopsy of skin shows a heavy band like infiltrate of small, irregularly contoured lymphocytes embedded in coarse, horizontally-disposed collagen fibers. There is continuous haphazard epidermotropism suggesting patch/plaque mycosis fungoides. Case 1 - Figure 1B - The punch biopsy of skin shows a heavy band like infiltrate of small, irregularly contoured lymphocytes embedded in coarse, horizontally-disposed collagen fibers. There is continuous haphazard epidermotropism suggesting patch/plaque mycosis fungoides. Case 1 - Figure 1C - The punch biopsy of skin shows a heavy band like infiltrate of small, irregularly contoured lymphocytes embedded in coarse, horizontally-disposed collagen fibers. There is continuous haphazard epidermotropism suggesting patch/plaque mycosis fungoides. Case 1 - Figure 2A Case 1 - Figure 2B Case 1 - Figure 3A - The punch biopsy specimens show a band like infiltrate of small lymphocytes with a directed pattern epidermal migration into an epidermis surmounted by parakeratotic mounds of plasma-containing scale crust. There is prominent involvement of the adnexal structures, which are sites of maximal antigenic processing. This is the classical pattern of lymphomatoid hypersensitivity. Both punch biopsy specimens show a band like infiltrate of small lymphocytes with a reduction on cell density over the orginal biopsy, in concert with discontinuous haphazard epidermal migration. Case 1 - Figure 3B - The punch biopsy specimens show a band like infiltrate of small lymphocytes with a directed pattern epidermal migration into an epidermis surmounted by parakeratotic mounds of plasma-containing scale crust. There is prominent involvement of the adnexal structures, which are sites of maximal antigenic processing. This is the classical pattern of lymphomatoid hypersensitivity. Case 1 - Figure 4 - Clinical examination shows no evidence of the rash at the truncal site of former maximal involvement. Discussion Clinical The concept of the lymphomatoid drug eruption, or drug-associated pseudomycosis fungoides (pseudo-MF), was first recognized when Dilantin was linked to lesions that clinically resembled MF [9]. Subsequent reports have associated similar lesions with intake of other drugs including phenothiazines, antihistaminics, anxiolytics, antidepressants, barbiturates, b-blockers, calcium channel blockers and ACE inhibitors [3, 4, 6, 10] . While the prototypic reaction pattern for the lymphomatoid drug response was one which resembled MF [9], other patterns since recognized include lymphocytoma cutis, follicular mucinosis, and atypical pigmentary purpura [2, 5] . The cumulative and/or synergistic effect of polypharmacy on immune dysregulation in the propagation of these eruptions has been proposed [7], as the various implicated drugs all have immune dysregulating properties (see Table 1), Table 1 - Drugs implicated in pseudolymphomata ACE inhibitors Alpha antagonists Anticonvulsants Antidepressants Benzodiazepines Beta blockers Calcium channel blockers H1 Antagonists H2 Antagonists Lithium Lipid lowering agents Non-steroidal anti-inflammatory agents Penothiazines Sex steroids Legend ACE - Angiotensin converting enzyme H1 - Histamine receptor, type 1 H2 - Histamine receptor, type 2 including in vitro effects on lymphocyte function such as promotion of lymphoid mitogenesis and suppression of T-suppressor function. We hold that any patient who develops an atypical lymphoid infiltrate while ingesting one or more of the aforementioned agents should have a drug-based etiology excluded before being held to have a cutaneous lymphoma. Exacerbation of pre-existing MF by fluoxetine has been shown [12] and it has been our experience that patients with malignant lymphoma who receive drugs from the aforementioned classes may transiently improve when the drug or drugs are withdrawn, only to subsequently relapse. It is possible that patients with lymphomatoid drug eruptions are not only those who receive immune-dysregulating agents, but also those with underlying endogenous immune dysregulation due to CTD, lymphoreticular neoplasia, HIV infection, and visceral malignancies [6]. Histopathology In the setting of plaques clinically resembling MF, biopsies generally show a superficial, band-like lymphocytic infiltrate with variable epitheliotropism typically directed to sites of antigenic processing such as suprapapillary plates, acrosyringea and hair follicles: Table 2 - Histologic Features of the Lymphomatoid Hypersensitivity Reaction The MF-like pattern of lymphomatoid hypersensitivity Features common to both MF and lymphomatoid hypersensitivity Infiltration of the epidermis by mildly atypical lymphocytes or by cells with a comparable cytomorphology to the small- and intermediate-sized atypical dermal lymphocytes. Sezary cells but No Pautrier's microabscesses Directed pattern of epidermal infiltration with maximal involvement of suprapapillary plates/adnexae* Vesiculation* Papillary dermal edema* Vascular fibrin deposition* (*characteristic of delayed type hypersensitivity) Dermal Sezary cells Atrophy Vascular ectasia Dermal fibrosis, either vertical or laminated In our hands, the intraepidermal lymphoid populace has proven to be less atypical than the dermal lymphocytes, which include transformed cells [2]. Expressing the same observation in another way, investigators from the European Organization for Research and Treatment of Cancer (EORTC) compared the nuclei of the intraepidermal lymphoid populace to the nuclear diameter of keratinocytes; lymphoid diameters similar to adjacent keratinocytes were found to be reasonably specific for MF, while in those cases that represented MF mimics, papillary dermal fibrosis or significant numbers of dermal 'blast-like' cells were seen [11]. Eosinophils and plasma cells are also common in our hands. Papular lesions of drug-induced pseudolymphoma may show a lymphomatoid vascular reaction, namely dense angiocentric atypical lymphoid infiltrates associated with variable luminal and mural fibrin deposition and ischemic epidermal alterations. Other light microscopic correlates of papular pseudolymphoma include follicular mucinosis accompanied by exocytosis of atypical lymphocytes including those with cerebriform nuclear contours. When nodules are biopsied, a diffuse and/or follicular lymphocytoma cutis pattern is often observed. The final pattern resembles the atypical pigmentary purpura-like reaction of MF [2]. Drug-associated atypical pigmentary purpura is difficult to distinguish from the atypical pigmentary purpura which occurs as a manifestation of MF, perhaps the most useful discriminating light microscopic feature favoring the latter being the presence of a more atypical intraepidermal lymphoid populace relative to the dermal-based lymphoid infiltrate. Drug-induced reversible lymphoid dyscrasia We have encountered cases of atypical T-cell lymphoid hyperplasia in the setting of therapy with drugs of the prototypic classes associated with atypical lymphoid proliferations in which we explored phenotypic anomalies using antibodies to CD3/4/5/7/8 and CD62L and sought T-cell receptor (TCR) gene rearrangements by a polymerase chain reaction (PCR) methodology. The lymphoid infiltrates showed reproducible CD7 and/or CD62 deletion in concert with T cell clonality, findings cognate to CTCL, but resolved or improved substantially following drug modulation. We hypothesize that the infiltrates represent an unrepressed expansion of CD7/CD62-negative memory T-lymphocytes in response to antigenic triggers. We propose the term drug associated reversible lymphoid dyscrasia for this entity [8]. References Crowson AN, Magro CM. Cutaneous pseudolymphoma : A review. Fitzpatrick's Journal of Clinical Dermatology 1995;3:43-55. Crowson AN, Magro CM, Zahorchak R Atypical pigmentary purpura : A clinical, pathological and genotypic study of 40 cases Hum Pathol 1999;30:1104-12. Crowson AN, Magro CM. Antidepressant therapy : A possible cause of atypical cutaneous lymphoid hyperplasia. Arch Dermatol 1995;131:925-929. Crowson AN, Magro CM. The pathology of cutaneous drug eruptions (monograph). St. Louis: CV Mosby. Current Problems in Dermatology 2002;14(4):117-146. Crowson AN, Brown TJ, Magro CM. Progress in the understanding of the pathology and pathogenesis of cutaneous drug eruptions: implications for management. Am J Clin Dermatol 2003;4:407-428. Magro CM, Crowson AN. Drugs with antihistaminic properties as a cause of atypical cutaneous lymphoid infiltrates. J Am Acad Dermatol 1995;32:419-28. Magro CM, Crowson AN. Drug-induced immune-dysregulation as a cause of atypical cutaneous lymphoid infiltrates : A hypothesis. Hum Pathol 1996;26:125-132. Magro CM, Crowson AN, Kovatich AJ, Burns F. Drug-induced reversible lymphoid dyscrasia: A clonal lymphomatoid dermatitis of memory and activated T cells. Hum Pathol 2003;34:119-129. Rijlaarsdam U, Scheffer E, Meijer CJLM, Kruyswijk MRJ, Willemze R. Mycosis fungoides-like lesions associated with phenytoin and carbamazepine therapy. J Am Acad Dermatol 1991;24:216-222. Rijlaarsdam JU, Scheffer E, Meijer CJLM, Willemze R. Cutaneous pseudo-T-cell lymphomas: A clinicopathologic study of 20 patients. Cancer 1992;69:717-724. Santucci M, Biggeri A, Feller AC, Massi D, Burg G. Efficacy of histologic criteria for diagnosing early mycosis fungoides. Am J Surg Pathol 2000;24:40-50. Vermeer MH,Willemze R. Is mycosis fungoides exacerbated by fluoxetine? J Am Acad Dermatol 1996;35:635-6.