THERAPY

Low-dose oral bexarotene in combination withlow-dose interferon alfa in the treatment of

cutaneous T-cell lymphoma: Clinical synergismand possible immunologic mechanisms

Karen S. McGinnis, MD, Jacqueline M. Junkins-Hopkins, MD,Glen Crawford, MD, Michael Shapiro, MD, Alain H. Rook, MD, and

Carmela C. Vittorio, MDPhiladelphia, Pennsylvania

Background: For nearly 2 decades clinicians have been treating cutaneous T-cell lymphoma (CTCL) withregimens that combine interferon alfa with retinoid compounds. In December 1999 a new retinoid,bexarotene, was approved by the US Food and Drug Administration for the treatment of CTCL. At themanufacturer’s recommended dose of bexarotene (300 mg/m2 of body surface area), it has proven to be ahighly effective therapy for all stages of CTCL. Nevertheless, this dose is typically associated with adverseeffects including severe hyperlipidemia. Furthermore, there appears to be no standardization of dosingamong physicians who treat CTCL.

Observations: We present 3 representative patients, 2 with erythrodermic CTCL and 1 with follicularmycosis fungoides, who experienced the rapid clearing of skin disease while being treated with acombination of low-dose bexarotene and low-dose recombinant interferon alfa.

Conclusions: Combining low-dose bexarotene with low-dose interferon alfa was well tolerated and led torapid improvement in our patients. We review the clinical and biologic basis for this approach. (J Am Acad

Dermatol 2004;50:375-9.)

F or nearly 2 decades clinicians have been treat-ing cutaneous T-cell lymphoma (CTCL) withregimens that combine interferon alfa with

retinoid compounds.1-8 In December 1999 a newretinoid, bexarotene, was approved by the Food andDrug Administration for the treatment of CTCL. Atthe manufacturer’s recommended dose of bexaro-tene (300 mg/m2 of body surface area), it has provento be a highly effective therapy for all stages ofCTCL. Nevertheless, this dose is typically associatedwith adverse effects including severe hyperlipid-emia.9,10 Furthermore, there appears to be no stan-dardization of dosing among physicians who treat

From the Department of Dermatology, University of Pennsylvania.a

Funding sources: None.Conflicts of interest: None identified.Reprint requests: Carmela C. Vittorio, MD, University of

Pennsylvania, School of Medicine, Department of Dermatology,2 Maloney, 3600 Spruce St, Philadelphia, PA 19104-4283. E-mail:carmela.vittorio@uphs.upenn.edu.

0190-9622/$30.00© 2004 by the American Academy of Dermatology, Inc.doi:10.1016/j.jaad.2003.10.669

CTCL. Thus, it has been our approach to use bex-arotene in low doses (100 mg/m2) in combinationwith low doses of interferon alfa. Although this isnot the first report of using bexarotene in combina-tion with other biologic agents, it is unique in thatnone of our patients required more than 150 mgdaily to achieve a major response.11 It is postulated,therefore, that therapeutic synergism occurred withthe combination therapy.

In this report we present 3 representative pa-tients, 2 with erythrodermic CTCL and 1 with follic-ular mycosis fungoides, who experienced the rapidclearing of skin disease while being treated with acombination of low-dose bexarotene and low-doserecombinant interferon alfa. Furthermore, we dis-cuss the immunologic mechanisms underlying theefficacy of these drugs and propose possible expla-nations for the observed synergy.

CASE REPORTSCase 1

An 85-year-old Caucasian man presented with a6-month history of erythroderma and severe pruritus

375

376 McGinnis et al J AM ACAD DERMATOL

MARCH 2004

that had responded minimally to fluocinolone ace-tonide 0.025% ointment and UVB phototherapy 3times weekly. On examination he was found to haveprofound total body erythroderma with scale (Fig 1,A). He did not have palpable cervical, axillary, oringuinal lymphadenopathy. The peripheral bloodrevealed a normal flow cytometry pattern and buffycoat analysis documented a Sezary cell count of 1%.Serum lactate dehydrogenase and a complete bloodcell count were both within normal limits. A skinbiopsy specimen was consistent with mycosis fun-goides. A diagnosis of erythrodermic CTCL wasmade and interferon alfa-2b was started at 1.2 mil-lion U subcutaneously 3 times per week, and wasgradually increased over a period of 5 weeks to 2.4million U subcutaneously 3 times per week. Oralbexarotene (150 mg/d) was also started. Mild centralhypothyroidism, manifested by low free T4 level,occurred 2 months after initiation of the bexaroteneand was treated with levothyroxine.13 Atorvastatin(10 mg/d) was administered prophylactically to pre-vent bexarotene-induced hyperlipidemia. Within 1month of beginning bexarotene and interferon ther-apy, the patient experienced a marked decrease inerythema and scale, with complete clearing of boththe erythroderma and pruritus at a 6-month follow-up. At that time the bexarotene was discontinuedwhereas the interferon was decreased to 1.8 millionU 3 times per week. At a 16-month follow-up hisskin remained clear and his interferon was stopped

Fig 1. Front of chest before (A) and after ((A) and complete clearance of erythema (B

(Fig 1, B). He has remained in remission and off alltherapy for the past 30 months.

Case 2A 56-year-old Caucasian man presented with

erythroderma and severe pruritus. The patient de-scribed the symptomatic onset in 1967 of scaling anderythema of the face and hands, gradually progress-ing over the course of 30 years. All previous biopsyspecimens and laboratory investigations were non-contributory; however, clinical suspicion generatedthe presumptive diagnosis of mycosis fungoides.Prior treatments included topical steroid prepara-tions, topical nitrogen mustard, topical carmustine,psoralen-UVA therapy, and etretinate. There wasonly a transient response to these modalities. Phys-ical examination at presentation revealed near-con-fluent erythroderma involving close to 80% of thebody surface area, and localized, indurated, scalingplaques, with sparing of the lower aspect of hisabdomen and back. There was no palpable cervical,axillary, or inguinal lymphadenopathy. Laboratoryinvestigation revealed normal liver function studies,including lactate dehydrogenase, and normal com-plete blood cell counts. Skin biopsy specimen wasconsistent with mycosis fungoides. On buffy coatanalysis the Sezary cell count was 4% to 8%. Subcu-taneous interferon alfa-2b was started at 1.8 millionU 3 times weekly. This resulted in prominent areasof clearing on his arms and abdomen, and a dra-

tment demonstrating diffuse erythroderma

B) trea).

McGinnis et al 377J AM ACAD DERMATOL

VOLUME 50, NUMBER 3

matic reduction in pruritus. However, the patient’serythroderma and pruritus recurred and worsenedduring the following 16 months, despite a gradualincrease in the interferon dose to 6 million U everyother day. At this time, the erythroderma was nearlyconfluent, involving close to 100% of his body sur-face area. Oral bexarotene at 150 mg daily wasadded and the interferon was decreased to 4 millionU 3 times weekly. Simvastatin (20 mg/d) was insti-tuted to correct bexarotene-induced hyperlipidemia.Six months after the institution of bexarotene ther-apy, the patient demonstrated a dramatic decrease inscaling and pruritus. At 9-month follow-up he con-tinues on the same pharmacologic regimen withonly faint erythema over the trunk and extremitiesnoted. Repeated Sezary cell counts have been in the3% range.

Case 3A 62-year-old Caucasian man presented with a

5-year history of a pruritic rash that began on histhighs and had been treated with moisturizer. Onexamination he was found to have widespread pap-ules and plaques consistent with CTCL. He did nothave palpable cervical, axillary, or inguinal lymph-adenopathy. A skin biopsy specimen taken from theleft thigh was read as folliculocentric CTCL. Theperipheral blood revealed a normal flow cytometrypattern and buffy coat analysis documented a Sezarycell count of 3%. Lactate dehydrogenase and a com-plete blood cell count were both within normallimits. A diagnosis of folliculocentric CTCL wasmade and interferon alfa-2b was started at 2.0 mil-lion U subcutaneously 3 times per week, oral bex-arotene (150 mg/d), and fenofibrate (134 mg/d).Levothyroxine was later added to correct bexaro-tene-induced central hypothyroidism. Within 1month, the patient exhibited a dramatic decrease inthe number of papules and plaques. Within 4months of beginning therapy there were only 3 faintpatches and a few papules remaining. At 21-monthfollow-up he continues on the same pharmacologicregimen with only a few areas of follicular involve-ment on the legs noted. His blood remains clear ofdisease.

DISCUSSIONCTCL is a clonal proliferative disease of skin-

homing helper T lymphocytes that express the T-helper 2 phenotype.12 Prognosis can range fromexcellent (stage IA, equal to age-matched controlsubjects) to poor (stage IV, 5-year survival of�40%).13,14 All of our patients would be consideredto have a poor prognosis; patients No. 1 and No. 2on the basis of erythroderma and patient No. 3 onthe basis of the folliculocentric nature of his dis-

ease.15 However, each patient responded rapidlyand completely to our combined, low-dose regimen.None were cultured or treated for Staphylococcus.

Retinoids and interferons have been used in thetreatment of CTCL for almost 20 years and may worktogether by several different mechanisms. The first isby simultaneously increasing type 1 helper T-cellcytokines and decreasing type 2 helper T-cell cyto-kines. Indeed, certain retinoids, including all trans-retinoic acid and 13-cis-retinoic acid, have beenshown to enhance interferon gamma and IL-12 pro-duction by peripheral blood mononuclear cells,whereas interferon alfa has been shown to inhibitIL-4 and IL-5 production by T cells of patients withSezary syndrome.16,17 Furthermore, both retinoicacid18,19 and interferons20,21 have been shown toinduce cell-mediated cytoxicity and stimulate natu-ral killer activity in vivo. Additional explanations fortheir apparent synergistic antineoplastic effect in-clude a combined antiproliferative effect22 and anti-angiogenic properties.23

Bexarotene is a new type of retinoid that is ahighly selective agonist for the retinoid X receptor(RXR).24 Bexarotene’s affinity for the RXR is in con-trast to traditional retinoids, which bind preferen-tially to the retinoic acid receptor.25-27 Both retinoicacid receptors and RXRs belong to the family ofnuclear hormone receptors, which include the thy-roxine receptor, peroxisomal proliferator-activatedreceptor (PPAR), and the vitamin D receptor.28,29

However, of this family, only RXRs can dimerizewith other nuclear hormone receptors to form het-erodimers.30 These heterodimers may then influencegene expression in 2 ways: direct binding to DNAwith transactivation, and inhibition or competitionwith other nuclear transcription factors.24,31

Of particular interest with respect to bexarotene’sefficacy in CTCL is the RXR-PPAR dimer. PPARs arewell known for their role in lipid metabolism, butmore recently have been shown to be importantregulators of epidermal32-34 and immune system35-36

function. For example, Clark et al36 found that PPAR-gamma agonists blocked murine-activated helper T-cell clonal proliferation while inhibiting IL-2 secre-tion. Although the inhibition of IL-2 would seem tobe detrimental with respect to immunomodulatorytherapy for CTCL, it is possible that a similar down-regulation of oversecreted T-helper type 2 cyto-kines, such as IL-4, may occur by the same nuclearmechanisms.24

Another possible role for the RXR-PPAR dimercould be the alteration of lymphocyte traffickingthrough the skin by down-regulation of E-selectin.E-selectin is a receptor on endothelial cells that isspecific for cutaneous lymphocyte antigen, which is

378 McGinnis et al J AM ACAD DERMATOL

MARCH 2004

typically found on skin-trafficking T cells includingthe malignant T cells in CTCL.37 In addition, RXR-PPAR heterodimers inhibit up-regulation of otherendothelial cell activation molecules and decreaseendothelial cell secretion of inflammatory media-tors.35,38 Therefore, bexarotene may confer sometherapeutic benefit simply by decreasing inflamma-tion and, thus, lymphocyte trafficking within theskin.

The RXR-PPAR dimer may also be involved in theability of RXR agonists to suppress proliferation andpromote apopotosis. Indeed, Zhang et al39 foundthat treatment of CTCL cell lines at clinically relevantconcentrations of bexarotene led to apoptosis. How-ever, the question remains: do biologic mechanismsexist by which interferon alfa specifically contributesto or augments the RXR-PPAR mediated pathways,thereby explaining the observed synergy in vivo? Orare the mechanisms of synergy more akin to thoseinvolved in the traditional retinoid/interferon com-binations discussed previously?

With respect to interferon and the RXR-PPARpathways, Tan et al40 showed that interferon gammainduced keratinocyte differentiation by a cytokine-dependent cell differentiation pathway that requiredup-regulation of the PPAR �/� gene.

With regard to specific immune system interac-tions, Chinetti et al41 demonstrated that PPARgamma ligands induce apoptosis of macrophagesactivated with interferon gamma by negative inter-ference with the antiapoptotic nuclear factor � Bsignaling pathway. These data suggest that perhapsinterferons may serve to “prime” the malignant T cellfor exposure to the RXR-PPAR heterodimer, whichcan then more effectively induce apoptosis.

Finally, it is interesting that two of our patientswere simultaneously treated with a member of thestatin family, which have also been shown to acti-vate PPAR receptors and abolish nuclear factor � Bactivity resulting in down-regulation of proinflam-matory cytokines and increased apoptosis in humanmonocytes.42,43

In summary, it appears that the combination ofinterferon alfa and bexarotene may modulate anti-tumor immunity and tumor regression through mul-tiple complex and interrelated pathways that couldlead to synergistic clinical effects. A similar rationalefor combined regimens that include bexarotene andinterferon with photopheresis, other cytokines, orpsoralen-UVA also deserves future exploration.Thus, there is a substantial need for controlled trialsthat explore the use of combined immunomodula-tory regimens in the treatment of CTCL.

We thank William Witmer for his assistance with thephotographic material.

REFERENCES1. Thestrup-Pedersen K, Hammer R, Kaltoft K, Sogaard H, Zachariae

H. Treatment of mycosis fungoides with recombinant interfer-on-alpha-2a alone and in combination with etretinate. Br J Der-matol 1988;118:811-8.

2. Braathen LR, McFadden N. Successful treatment of mycosis fun-goides with the combination of etretinate and human recombi-nant interferon alfa-2a. J Dermatol Treat 1989;1:29-32.

3. Knobler RM, Trautinger F, Radaszkiewicz T, Kokoschka EM,Micksche M. Treatment of cutaneous T-cell lymphoma with acombination of low-dose interferon alfa-2b and retinoids. J AmAcad Dermatol 1991;24:247-52.

4. Dreno B, Claudy A, Meynadier J, Verret JL, Souteyrand P,Ortonne JP, et al. The treatment of 45 patients with cutaneousT-cell lymphoma with low doses of interferon-alfa 2a and etreti-nate. Br J Dermatol 1991;125:456-9.

5. Altomare GF, Capella GL, Pigatto PD, Finzi AF. Intramuscular lowdose alpha-2b interferon and etretinate for treatment of myco-sis fungoides. Int J Dermatol 1993;32:138-41.

6. Dreno B. Roferon A (interferon alpha 2a) combined with Tigison(etretinate) for treatment of cutaneous T cell lymphomas. StemCells 1993;11:269-75.

7. Torii H, Kaneko T, Matsuyama T, Nakanishi H, Harada S. Interfer-on-alpha (IFN-alpha) and etretinate in the treatment of mycosisfungoides. J Dermatol 1994;21:767-70.

8. Stadler R, Otte HG, Luger T, Henz BM, Kuhl P, Zwingers T, et al.Prospective randomized multicenter clinical trial on the use ofinterferon alpha-2a plus acitretin versus interferon alpha-2aplus PUVA in patients with cutaneous T-cell lymphoma stages Iand II. Blood 1998;92:3578-81.

9. Duvic M, Martin A, Kim Y, Olsen E, Wood GS, Crowley CA, et al.Phase 2 and 3 clinical trial for oral bexarotene (Targretin cap-sules) for the treatment of refractory or persistent early-stagecutaneous T-cell lymphoma. Arch Dermatol 2001;137:581-93.

10. Duvic M, Hymes K, Heald P, Breneman D, Martin AG, MyskowskiP, et al. Bexarotene is effective and safe for treatment of refrac-tory advanced-stage cutaneous T-cell lymphoma: multinationalphase II-III trial results. J Clin Oncol 2001;19:2456-71.

11. Talpur R, Ward S, Apisarnthanarax N, Breuer-McHam J, Duvic M.Optimizing bexarotene therapy for cutaneous T-cell lymphoma.J Am Acad Dermatol 2002;47:672-84.

12. Diamandidou E, Cohen PR, Kurzrock R. Mycosis fungoides andSezary syndrome. Blood 1996;88:2385-409.

13. Foss FM, Sausville EA. Prognosis and staging of cutaneous T celllymphoma. Hematol Oncol Clin North Am 1995;9:1011-9.

14. Zackheim HS, Amin S, Kashani-Sabet M, McMillan A. Prognosisin cutaneous T-cell lymphoma by skin stage: long term survivalin 489 patients. J Am Acad Dermatol 1999;40:418-25.

15. van Doorn R, Scheffer E, Willemze R. Follicular mycosis fun-goides, a distinct disease entity with or without associated fol-licular mucinosis. Arch Dermatol 2002;138:191-8.

16. Fox FE, Kubin M, Cassin M, Niu Z, Trinchieri G, Cooper KD, et al.Retinoids synergize with interleukin-2 to augment IFN-gamma and interleukin-12 production by human peripheralblood mononuclear cells. J Interferon Cytokine Res 1999;19:407-15.

17. Suchin KR, Cassin M, Gottleib SL, Sood S, Cucchiara AJ,Vonderheid EC, et al. Increased interleukin 5 production in eo-sinophilic Sezary syndrome: regulation by interferon alfa andinterleukin 12. J Am Acad Dermatol 2001;44:28-32.

18. Lotan R, Dennert G. Stimulatory effects of vitamin A analogueson induction of cell mediated cytotoxicity in vivo. Cancer Res1979;39:55-8.

19. Denert G, Lotan R. Effects of retinoic acid on the immune

McGinnis et al 379J AM ACAD DERMATOL

VOLUME 50, NUMBER 3

system: stimulation of T killer cell induction. Eur J Immunol1978;8:23-9.

20. Herberman RB, Ortaldo JR, Mantovani A, Hobbs DS, Kung HF,Pestka S. Effect of recombinant interferon on cytotoxic activityof NK cells and monocytes. Cell Immunol 1982;67:160-7.

21. Goldfarb RH, Herberman RB. Natural killer cell reactivity: regula-tory interactions among phorbol ester, interferon, cholera toxin,and retinoic acid. J Immunol 1981;126:2129-35.

22. Lancillotti F, Giandomenico V, Affabris E, Fiorucci G, Romeo G,Rossi GB. Interferon alpha-2b and retinoic acid combined treat-ment affects proliferation and gene expression of human cervi-cal carcinoma cells. Cancer Res 1995;55:3158-64.

23. Majewki S, Szmurlo A, Marczak S, Jablonska S, Bollag W. Syner-gistic effect of retinoids and interferon alpha on tumor-inducedangiogenesis: anti-angiogenic effect of HPV-harboring tumor-cell lines. Int J Cancer 1994;57:81-5.

24. Cheng SX, Kupper T. A new rexinoid for cutaneous T-cell lym-phoma. Arch Dermatol 2001;137:649-52.

25. Giguere V, Ong ES, Segui P, Evans RM. Identification of a recep-tor for the morphogen retinoic acid. Nature 1987;330:624-9.

26. Petkovich M, Brand NJ, Krust A, Chambon P. A human retinoicacid receptor which belongs to the family of nuclear receptors.Nature 1987;330:444-50.

27. Mangelsdorf DJ, Ong ES, Dyck JA, Evans RM. A nuclear receptorthat identifies a novel retinoic acid response pathway. Nature1990;345:224-9.

28. Evans RM. The steroid and thyroid receptor superfamily. Science1988;240:889-95.

29. Lotan R, Clifford JL. Nuclear receptors for retinoids: mediators ofretinoid effects on normal and malignant cells. Biomed Pharma-cother 1991;45:145-56.

30. Manglesdorf DJ, Evans RM. The RXR heterodimers and orphanreceptors. Cell 1995;83:841-50.

31. Szonday A, Reichert U, Fesus L. Retinoic acids regulate apoptosisof T lymphocytes through an interplay between RAR and RXRreceptors. Cell Death Differ 1998;5:4-10.

32. Komuves LG, Hanley K, Lefebvre A-M, Man MQ, Ng DC, Bikle DD,et al. Stimulation of PPARalpha promotes epidermal keratino-cyte differentiation in vivo. J Invest Dermatol 2000;115:353-60.

33. Komuves LG, Hanley K, Man M-Q, Elias PM, Williams ML,

Feingold KR. Keratinocyte differentiation in hyperproliferative

epidermis: topical application of PPAR alpha activators restorestissue homeostasis. J Invest Dermatol 2000;115:361-7.

34. Hanley K, Jiang Y, He SS, Friedman M, Elias PM, Bikle DD, et al.Keratinocyte differentiation is stimulated by activators of thenuclear hormone receptor PPAR alpha. J Invest Dermatol 1998;110:368-75.

35. Yang XY, Wang LH, Chen T, Hodge DR, Resau JH, DaSilva L, et al.Activation of human T lymphocytes is inhibited by peroxisomeproliferator-activated receptor gamma (PPAR gamma) agonist.J Biol Chem 2000;275:4541-4.

36. Clark RB, Bishop-Bailey D, Estrada-Hernandez T, Hla T,Puddington L, Padula SJ. The nuclear receptor PPAR gammaand immunoregulation: PPAR gamma mediates inhibition ofhelper T-cell responses. J Immunol 2000;164:1364-71.

37. Nawa T, Nawa MT, Cai T, Zhang C, Uchimura I, Narumi S, et al.Repression of TNF-alpha-induced E-selectin expression by PPARactivators: involvement of transcriptional repressor LRF-1/ATF3.Biochem Biophys Res Commun 2000;275:406-11.

38. Marx N, Mach F, Sauty A, Leung JH, Sarafi MN, Ransohoff RM, etal. Peroxisome proliferator-activated receptor gamma activa-tors inhibit IFN gamma-induced expression of the T cell activeCXC chemokines IP-10, Mig and I-TAC in human endothelialcells. J Immunol 2000;164:6503-8.

39. Zhang C, Hazarika P, Ni X, Weidner D, Duvic M. Induction ofapoptosis by bexarotene in cutaneous T-cell lymphoma cells:relevance to mechanism of therapeutic action. Clin Cancer Res2002;8:1234-40.

40. Tan NS, Michalik L, Noy N, Yasmin R, Pacot C, Heim M, et al.Critical roles of PPAR beta/delta in keratinocyte response to in-flammation. Genes Dev 2001;15:3263-77.

41. Chinetti G, Griglio S, Antonucci M, Torra IP, Delerive P, Majd Z, etal. Activation of proliferator-activated receptors alpha andgamma induces apoptosis of human monocyte-derived macro-phages. J Biol Chem 1998;273:25573-80.

42. Grip O, Janciauskiene S, Lindgren S. Atorvastatin activates PPAR-gamma and attenuates the inflammatory response in humanmonocytes. Inflamm Res 2002;51:58-62.

43. Zelvyte I, Dominaitiene R, Crisby M, Janciauskiene S. Modulationof inflammatory mediators and PPAR-gamma and NF-kappa-Bexpression by pravastatin in response to lipoproteins in human

monocytes in vitro. Pharmacol Res 2002;45:147-54.