Isotretinoin and Cutaneous Helper T-Cell Lymphoma (Mycosis Fungoides)

s

Isotretinoin and Cutaneous Helper T-Cell Lymphoma (Mycosis Fungoides) John F. Kessler, MD; Stephen E. Jones, MD; Norman Levine, MD; Peter J. Lynch, MD; Ann Rohman Booth, RN; Frank L. Meyskens, Jr, MD \s=b\ Retinoids, including isotretinoin, have demonstrated antiproliferative and antineoplastic activity in laboratory and clinical trials.In a phase II trial, 25 patients with extensive mycosis fungoides were evaluated for response to isotretinoin.There was a 44% (11 patients) objective clinical response rate with three clinical com- plete responses without concomitant evidence of pathologic clearing of the disease.An additional 24% (six patients) showed a minor degree of clinical improvement.The median time to response was two months (range, 0.5 to eight months) and the median response duration was eight months or longer (range, one to 25 months).Chron- ic toxic reactions consisted primarily of drying of the skin and mucous membranes and resulted in dose reduction in the majority of patients.It is concluded that isotretinoin produces significant clinical benefit to some patients with mycosis fungoides.
(Arch Dermatol 1987;123:201-204) Vitamin A and its analogues, the retinoids, have demonstrated antiproliferative activity in a • number of malignant and premalignant conditions .15In the laboratory, these agents can block phenotypic cell transformation induced by radiation, chemical agents, and growth factors."7Additionally, they have induced cellular maturation in promyelocytic leuke¬ mia and neuroblastoma cells in culture89 and have demonstrated the capability to modulate a number ' of immunologie functions, particularly cell-mediated Accepted for publication July 3, 1986.From the Sections of Hematology/Oncology (Drs Kessler, Jones, and Meyskens and Ms Booth) and Dermatology (Drs Levine and Lynch), Department of Internal Medicine, and Arizona Can- cer Center, University of Arizona, Tucson.Dr Jones is now with the Charles A. Sammons Cancer Center, Baylor University Medi- cal Center, Dallas.cytotoxicity.1011These promising features have prompted clinical evaluation of these compounds in a variety of cancerous and precancerous lesions as well as in benign proliferative disorders.
For editorial comment see p 189.
Mycosis fungoides is an indolent cutaneous T-cell lymphoma for which a number of partially success¬ ful treatments have been used, including electron beam irradiation to the skin surface, topical mech- lorethamine, and oral methoxsalen plus ultraviolet A light (PUVA).We have previously reported our experience with isotretinoin in a broad phase II anticancer trial that included four patients with mycosis fungoides, all of whom achieved an objective response.12Because of this encouraging initial experi¬ ence, we have treated 21 additional patients with mycosis fungoides.In our total group there were three complete responses and eight additional par¬ tial responses.

PATIENTS AND METHODS
Patients with clinically and histologically documented mycosis fungoides or Sézary syndrome were eligible.Pathologic evaluation included light microscopy, electron microscopy, and immunopathologic study (indirect immu¬ nofluorescence with OKT3, OKT4, and OKT8) of skin biopsy specimens (performed by N.L. or P.J.L.).All biop¬ sies were performed and specimens reviewed by University of Arizona (Tucson) pathologists.All patients were required to have at least advanced plaque stage disease (Committee on Staging and Classification of Cutaneous T-Cell Lymphomas, skin stage T2 or greater).13No exclu¬ sions were made for lymph node or visceral involvement, previous treatment, or reduced performance status.Informed consent was obtained in all cases in a protocol approved by the University of Arizona Institutional Review Board.
tPUVA indicates oral methoxsalen plus ultraviolet A light.
examination, complete blood cell count and differential cell count, examination of the buffy coat by light microscopy for Sézary cells, routine fasting blood chemistry studies and lipid profile, retinoid blood levels, color photographs of skin involvement, and a baseline toxic reaction evaluation.Appropriate roentgenograms and scans were obtained only for suspected visceral involvement.Follow-up of patients was generally at monthly intervals.
In the first 16 patients, treatment consisted of isotreti¬ noin initiated at a dose of 2 mg/kg/d orally in single or divided doses.Reduction in dose was allowed in accordance with previously published toxicity scales.14Subsequent patients were started at lower doses (1.0 mg/kg/d).Treat¬ ment was continued until disease progression or intolera¬ ble toxic reactions ensued.Response durations were calcu¬ lated from the time of initial response.The clinical response criteria were as follows: A complete clinical response (CCR) was defined as complete disappearance of all lesions lasting at least four weeks.In those patients achieving a CCR, a biopsy specimen of the previously most-involved areas was obtained one month following a clinical complete remission.A partial response (CPR) was at least a 50% decrease in assessable lesions that lasted at least four weeks, while a minor response involved definite improvement but less than a CPR.Disease progression was characterized by the unequivocal appearance of new lesions or a greater than 25% increase in size of the old lesions.

RESULTS
Twenty-eight patients were entered in the study.Three were rendered ineligible because of a lack of follow-up information (one patient) or because of being reclassified on review as having the more aggressive entity, peripheral T-cell lymphoma (two patients).These two patients developed early viscer¬ al involvement and died shortly thereafter.The Sixty-two-year-old man with extensive plaque-stage mycosis fungoides over his entire body (his face was spared).Repre¬ sentative baseline photographs of his legs are shown at left.He was started on a regimen of isotretinoin, 1 mg/kg/d, and partial clinical response was noted after five months of therapy (center).Although his skin lesions subsequently completely .cleared and clinical complete response was obtained (as represented in photograph at right), biopsy specimen of previ¬ ously involved skin revealed residual atypical lymphocytes.His clinical response continues with low-dose (10 mg daily) main-" tenance ¡sotretinoin therapy for more than 21 months.characteristics of the 25 assessable patients are shown in Table 1.The mean age was 65 years, with a «.range of 39 to 93 years.Eighteen patients were male and seven were female.Skin involvement included advanced plaques composed of more than 10% of the skin surface in 15 patients, tumor nodules in three patients, and erythroderma in seven patients.Palpa¬ ble adenopathy was present in ten patients: four of these had lymph nodes that were soft and small.Two patients underwent lymph node biopsy and speci¬ mens from both revealed involvement with atypicala lymphocytes consistent with mycosis fungoides.No patient had detectable visceral involvement.Blood involvement with Sézary cells was found in five ' patients, with greater than 5% of circulating lym¬ phocytes in three patients.Seven patients had received one prior treatment modality, six individu-* als had received two, and ten had been treated with three or more modalities.Two patients had not received prior treatment.
There were 11 (44% ) objective clinical responses to isotretinoin.An example of an objective clinical response is shown in the Figure and described in the legend.Three CCRs were noted with total disappear-" anee of all visible skin lesions; however, random skin biopsy specimens of previously involved skin revealed residual atypical lymphocytes in the epidermis.Eight patients achieved CPRs.Objective responses occurred with advanced plaques (seven [43%] of 15) and with erythroderma (four of seven) < but not with tumor nodules (T3) (zero of three).
There were three responses noted among the nine patients who had received three or more prior treat¬ ments.Palpable adenopathy improved in the three responders with palpable nodes; however, two of these patients had small soft nodes, and the third had no residual adenopathy following biopsy.
In addition to the 11 patients with significant responses, there were six patients with minor responses.The improvement in these latter patients was characterized by reduction in pruritus and a decrease in the degree of plaque infiltration but with minor change in plaque area.Seven patients had no response to the retinoid.The time from initiation of treatment to clinical response (CCR and CPR) ranged from two weeks to eight months, with a median of two months.The median duration of response (CCR and CPR) was more than eight months, with a range of one to 25 months.Patients continued to receive isotretinoin as , long as stabilization or response was clinically evi¬ dent.

TOXIC REACTIONS
The toxic reactions that were encountered are shown in Table 2. Most problems involved drying of the skin and mucous membranes and were usually controlled with emollients or reductions of the reti¬ noid dose.Additional effects included mild to moder¬ ate fatigue, arthralgias, myalgias, minor mental status changes (usually irritability), and mild to moderate headache.
Individuals with initially elevated triglycéride levels experienced the largest changes.No symptomatic hypertriglyceridemia was encountered.There were no treatment-related changes in cholesterol level, liver enzymes, or renal function.One patient's serum creatinine concentration increased with treatment but was best ascribed to her underlying renal dis¬ ease.No hématologie toxic reactions were encoun¬ tered.Dosage adjustments were needed in most patients, usually due to skin and mucous membrane dryness or fasting hypertriglyceridemia.Of the 16 patients starting treatment with 2 mg/kg/d, 14 patients required dose reductions (usually to about 1 mg/ kg/d), 12 for mucous cutaneous side effects and two for markedly elevated hypertriglyceridemia.One patient tolerated an increase in dose to 3 mg/kg/d.
Of six patients starting treatment with less than 2 mg/kg/d, two patients required dose reductions and two tolerated increases.Responses were seen at all dose levels whether initiated at 2 mg/kg/d or less.Sustained responses as long as 15 months or more have been noted with doses as low as 10 to 20 mg daily in several patients whose plaques cleared at higher doses but who were unable to tolerate these higher levels because of continued cutaneous toxic reactions.

COMMENT
We have demonstrated that isotretinoin produces an objective clinical response rate of 44% (11 patients) with significant improvement in an addi¬ tional 24% (six patients) of patients with mycosis fungoides.Similar results have been noted by others using this retinoid and other derivatives as well.Warrell et al15 reported responses in three of seven patients treated with isotretinoin, and a report by the Scandinavian Mycosis Fungoides Study Group described objective responses in 19 of 24 patients treated with isotretinoin.16 Mahrle et al,17 using an arotinoid noted CCRs in three of five patients and one CPR.Arotinoid responses occurred in the plaque-and tumor-stage of the disease but not in one patient with Sézary syndrome.Claudy and Rou- chouse18 reported improvement in 11 patients with parapsoriasis en plaques treated with the ethyl ester retinoid derivative (Etretinate).18Pathologically, they noted clearing of intraepidermal infiltrates, including Pautrier microabscesses.These studies and our report substantiate that retinoids can sup¬ press or reverse the clinical manifestations of myco¬ sis fungoides.
As has been found with other dermatologie condi¬ tions, different retinoids may exhibit different degrees of activity in the treatment of specific dis¬ eases.A large number of retinoids have been developed using laboratory systems, many of which exhibit more favorable therapeutic indexes than isotretinoin.It remains to be determined whether isotretinoin is the most active retinoid in mycosis fungoides.
Retinoid treatment should probably not supplant such standard treatment methods as electron beam radiation therapy, a topical nitrogen mustard, or PUVA, as these have resulted in long-term disease control."20The majority of patients relapse and require further therapy, however, and retinoids may be a relatively nontoxic, relatively effective alterna¬ tive.Cytotoxic therapy, while demonstrating activity in this disease, is accompanied by considerable potentially debilitating or even life-threatening toxic reactions and is not considered curative.21 Other therapeutic strategies should also be explored.of retinoids with interferon, an active agent for the treatment of mycosis fun¬ goides,22 would be plausible, perhaps allowing dose reductions of both agents.Additionally, the use of retinoids in patients at high risk of relapse following electron beam radiation therapy (ie, advanced dis¬ ease) should be evaluated as a maintenance regimen, as has been successfully done with topical mechlore- thamine.1'Isotretinoin produces significant clinical response of mycosis fungoides in heavily pretreated patients.We recommend a starting dose of 1 mg/kg/d in two divided doses.In some cases cutaneous toxic reac- tions from the drug were initially difficult to distin¬ guish from progress or flare of the disease.In such situations the dose was decreased by 50% to deter¬ mine the appropriate interpretation.The role of isotretinoin in previously untreated patients remains to be determined, but we recommend that isotretinoin be used after more conventional simple measures, such as topical steroids and perhaps PUVA, have been exhausted.Alternatively, use of isotretinoin with topical steroids may be well worth considering.< This investigation was supported in part by grant CA 27502 from the National Cancer Institute, National Institutes of Health, Bethesda, Md, and Hoffmann-La Roche Inc, Nutley, NJ.

Table 2 .
-Side Effects in 25 Patients Treated With Isotretinoin