The treatment of S6zary syndrome R. K. Winkelmann, M.D. , Ph.D., Jose L. Diaz-Perez, M.D., and Stanislav A. Buechner, M.D.* Rochester, MN

The results of treatment programs were studied in forty patients with S6zary syndrome. Thirty-three patients died. The principal treatment was a low-dose chlorambucil and prednisone regimen. Patients so treated lived longer from diagnosis of erythroderma (median survival, 6.2 yr) than did patients on other regimens (median survival, 3.05 yr). The addition of x-ray or chemotherapeutic programs did not increase significant long-term benefit. Seven patients receiving chlorambucil and prednisone had remission (I yr or more); most patients had partial remission, but two patients did not have a response to this program. The complications of sepsis, progressive disease, and lymphoma occurred both in patients who were and in those who were not treated with chlorambucil and prednisone. We believe that a regimen of low-dose chlorambucil and prednisone is satisfactory treatment that can be the basis for comparison of new therapeutic approaches to S6zary syndrome. (J AM ACAD DERMATOL 10:1000-1004, 1984,)

The chronic erythroderma of S6zary syndrome resists standard therapy. Almost by definition, this T cell erythroderma does not respond to systemic prednisone therapy except for short periods. Short-term beneficial response of hospital bed rest and nursing care with topical steroids and wet dressings is almost immediately followed by flares of the erythroderma when the patient returns to the home and work environment. The recurrent flares and chronic course produce disability and discour- age both the patient and the physician.

We introduced low-dose chlorambucil and prednisone therapy and summarized our early ex- perience with this program in S6zary syndrome in 1974.1 We have now reviewed experience with forty patients and our results with several treatment

From the Department of Dermatology, Mayo Clinic and Mayo Foun- dation.

Accepted for publication Nov. 4, 1983.

Reprint requests to: Dr. R. K. Winkehnann, Department of Der- matology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.

*Visiting clinician from the Department of Dermatology, University of Basel, supported by a grant from the Swiss National Science Foundation.

1000

programs, Comparing the chlorambucil (Leukeran) and prednisone program with other treatments shows that, as measured by duration of life from the onset of erythroderma, this program has an advantage.

MATERIALS AND M E T H O D S

We reviewed and summarized the course and treat- ment of forty patients seen at the Mayo Clinic with the diagnosis of S6zary syndrome. Such patients by defi- nition have a chronic erythroderma, cycles of S6zary cells in the peripheral blood of 15% or more or over 1,000/ram 3 on a continuing basis, and usually skin his- tologic findings of subepidermal lymphocytic bands with varying numbers of cerebfiform of S6zary cells. Although a provisional diagnosis of S6zary syndrome can be made on histologic and clinical findings alone, all the patients in this series had chronically and consis- tently elevated numbers of S~zary cells in the periph- eral blood.

All treatment given after the diagnosis of S6zary syndrome had been made was reviewed and summa- rized (Table I). The standard dose of chlorambucil was 4 rag/day, and usually the prednisone was given as 20 mg in a single morning dose. One patient received 2 mg of chlorambucil to start, and two received 6 rag. One

Volume 10 Number 6 June, 1984

Treatment of Sdzary syndrome 1001

T a b l e I. Treatment programs for S6zary syndrome (40 patients)

I "' No. of patients

Chlorambucil 25 X-ray therapy 15 Psoralens and ultraviolet A 4 Leukapheresis 3 Chemotherapy 14

Single-agent 14 Polychemotherapy 4

Tilorone 3

patient received as much as 10 mg/day in a short course of 8 weeks for an exacerbation of the disease. Long- term administration of chlorambucil was 2 to 4 mg/ day, and administration of prednisone was decreased to maintenance dosage of 5 to 10 mg/day or discontinued as improvement permitted. In three patients, the pro- gram was discontinued because of lowered peripheral blood counts. Treatment also was stopped when remis- sion was achieved. Duration of therapy is shown in Fig. 1.

Roentgen therapy was given as a total of 300 rads (in air) to each of four anterior and four posterior body fields at 100 kV orthovoltage and half-value layer of 1 mm aluminum. Single fractions of 150 R were repeated on two occasions to each field. Fifteen patients received such x-ray treatment; nine were receiving chlorambucil and prednisone, and six were not. Five patients later treated with chlorambucil and prednisone began ther- apy with this x-ray regimen, and five patients never treated with chlorambucil and prednisone began with this x-ray program. Three patients had two such x-ray treatments at intervals greater than 1 year.

Four patients were treated with psoralens and ul- traviolet A (PUVA).

Drug treatment included application of nitrogen mustard to the skin (4 patients), single-agent chemo- therapy, and polychemotherapy. Single-agent chemo- therapy included weekly courses of methotrexate (4 patients) and Vinca alkaloids (2) and daily doses of cyclophosphamide (Cytoxan) (2), azathioprine (Im- uran) (1), and triethylenemelamine (1). The patient given azathioprine received it for 1 year with minimal effect. Four patients were treated with courses of polychemotherapy. Three had variations of the cy- clophosphamide, vincristine, prednisone, and procar- bazine regimen at other institutions. One patient, who was treated with bleomycin, doxorubicin (Adriamy-

lO

9

8

7

N o . e

patients 5 .4

2

O

1 2 3 4 5 6 7 fl

Years of theraDy

Fig. 1. Duration of chlorambucil and prednisone ther- apy in twenty-five patients with S6zary syndrome.

cin), cyclophosphamide, vincristine (Oncovin), predni- sone, and procarbazine (BACOPP) at our institution for 5 months and had no skin or visceral response, was later treated with chlorambucil and prednisone for 3 months without response.

Three patients were treated with tilorone in a pro- spective, phase II study that was reported. 2 Leukapher- esis in three patients was not done weekly or pursued vigorously enough to be evaluated. In one patient treated witti electron-beam therapy, multiple ulcer- ations developed in the skin before 1,000 rads was delivered.

Complete remission was defined as absence of clini- cal signs and symptoms and absence of circulating S6zary cells on repeated examinations. Partial remis- sion represented improvement in erythroderma, devel- opment of areas of normal skin, and decrease of circu- lating Sfzary cells to pre-S6zary syndrome levels (usu- ally 100-500/mm '~) or to normal.

RESULTS

The median age of the forty patients was 60.4 years at the onset of erythroderma. The age range was 26 to 82 years, but in most patients the disease developed after the age of 60. Twenty-six patients were men, and fourteen were women.

At the time of this report, thirty-three patients were dead and seven were living. The onset of the erythroderma was taken as a clear time indication for these patients. We summarized all the patients ' courses by the duration of life from that time ex- pressed as the mean and standard deviation, and we organized the patients who were alive and those who were dead according to the treatment given (Table II). The duration of the disease or of

1002 W i n k e l m a n n et al

Journal of the American Academy of

Dermatology

Table II. Results of S~zary syndrome therapy (40 patients)

No. of patients

Years (mean _ S.D.) from onset of

Erythroderma { Skin disease*

Alive 7 4.6 + 2.6 8.1 + 5.7 Dead 33

Chlorambucil and prednisone 20 6.2 -+ 2.6 8.3 _ 3.2 Plus x-ray 9 6.6 -+ 2.6 8.5 -+ 2.7 Plus chemotherapy 6 6.7 -+ 3.1 9.7 -+ 2.1

Without chlorambueil and prednisone 13 3.05 -+ 1.8 5.6 - 4.7 Plus x-ray 6 2.0 + 1.2 3.6 - 1.8 Plus chemotherapy 5 3.2 _ 2.3 4.9 -+ 4.1

*Duration from onset of skin disease to death or to present status; includes duration from diagnosis of erythroderma.

life until death was twice as long in patients treated with low-dose chlorambucil and predni- sone as in those not treated by this means. By including the extended period from onset of defi- nite skin disease, if any, preceding erythroderma, we did not find any change in the relationships in any group (Table II). Addition of x-ray therapy as the low-voltage, low-half-value-layer regimen did not change the results either with or without chlorambucil and prednisone. Similarly, no change in duration of life was produced by the single-agent chemotherapy or polychemotherapy used.

Seven patients were living at the time of this report, and three of them were in remission. In one, remission occurred spontaneously with local therapy, and in two, it was achieved with chlorambucil and prednisone. Four patients were in partial remission with chlorambucil and predni- sone therapy.

The median age for the group of twenty patients treated with chlorambucil and prednisone who died was 59.3 years, similar to the median age of the entire group of patients. All patients had a good initial response to the treatment. The re- sponse was not rapid. During the first 1 to 3 months, there was a reduction in pruritus. By 3 to 6 months, a change in the erythema and scaling had occurred.

Seven of our nineteen patients had complete remission for longer than 1 year, requiring only local lubrication or occasional application of topi- cal corticosteroids. Partial remission occurred in most of the rest of the patients. Chlorambucil and prednisone had no effect on two patients, who

died within 6 months of its use. Significant remis- sions lasting for any length of time did not occur with any of the other treatment programs.

Ten of the seventeen patients receiving pro- longed chlorambucil and prednisone therapy had significant and persistent decrease in the values of circulating S6zary cells. Within 6 months, the val- ues were at pre-S6zary syndrome levels (below 1,000/mm3), and usually thereafter percentages of 1 to 3 or absolute values of 100-300/ram '~ were common. We used the clinical status and the pe- ripheral S6zary cell counts to monitor the disease. Although biopsies had helped us monitor tilorone therapy responses in a prospective study, this ret- rospective review found that not enough later, repetitive biopsies had been done in this larger group to make evaluation possible.

Flares of S6zary syndrome occurred with irrita- tion, nervous tension, drug administration, ac- tivities such as painting and wallpapering, sun ex- posure, and contact dermatitis. The S6zary cell counts did not regularly reflect these flares of in- creased redness and symptoms. Flares of eryth- roderma occurred on lower, maintenance doses of chlorambucil and prednisone and as the patients increased activities. Such flares were handled by bed rest, local wet dressings and corticosteroid creams, and return to full oral therapy; this regi- men produced improvement again. Steadily in- creasing numbers of circulating S6zary cells or leukocytes, or both, during this regimen correlated with a poor prognosis and progressive disease. Increasing the chlorambucil dose at such times was of limited and short-term benefit.

The use of other therapies, such as PUVA, tar

Volume I0 Number 6 June, 1984

Treatment of Sdzary syndrome 1003

and ultraviolet light, tilorone, and chemotherapy, did not add to any specific advantage to our basic program. The effect of tilorone on S6zary and pre-S6zary cases has been reported elsewhere. 2 One of four patients treated with PUVA had sig- nificant improvement, but another had aggravation of the skin and a clear rise in peripheral circulating S6zary cells; the other two patients did not have improvement. Leukapheresis temporarily lowered S6zary cell counts, but it was not pursued vigor- ously as a weekly treatment in these patients. No single chemotherapeutic agent demonstrated any advantage by local or systemic route. In four pa- tients treated with local nitrogen mustard, therapy was usually stopped within 3 months because of irritation or lack of success. Polychemotherapy, given as a final treatment in three patients, led to improvement lasting only 1 month to 3 months. One patient, who had the BACOPP regimen for 5 months, did not have improvement.

The cause of death was established in thirty of the thirty-three patients who died. Because they were older patients, it is perhaps not surprising that many of them died of vascular disease or sep- sis. Five patients died of myocardial infarctions and three of cerebrovascular accidents. One pa- tient died of cardiac failure, and one died of a pulmonary embolus. Bronchopneumonia was pres- ent in two patients, and Pneumocystis carinii was identified in the patient with multiple myeloma. Eight patients died of sepsis, and four died with suppression of formed elements of the blood and a history compatible with chemotherapy overre- sponse. Four patients had progressive T cell dis- ease. Thirteen autopsy protocols revealed lym- phoma in three instances, lymphatic leukemia in one, Hodgkin's disease in one, and multiple my- eloma in one. One patient died of metastatic squamous cell carcinoma. These data were sum- marized elsewhere. :~

DISCUSSION

Treatment programs for S6zary syndrome have included single-agent chemotherapy, 4,~ polyche- motherapy, 6 and x-ray and electron treatment. 7 None of these has been considered successful or useful enough to be considered a best treatment.

A rare occurrence in S6zary syndrome is remis- sion with local skin care alone. In one patient in

this series, who had erythroderma, lymphadenopa- thy, and a high circulating S6zary cell count, re- mission developed without any more therapy. This outcome is so rare in our experience that one is led to believe that the patient might not have true S6zary syndrome despite the findings confirming the diagnosis and the duration of the disease (4 yr). The process certainly did not behave in a manner similar to that in our other thirty-nine pa- tients, who had a chronic course with continued flares of erythroderma or remissions induced by chlorarnbucil and prednisone.

In our original reported series of S6zary cases,1 a number of patients had "chemotherapy" deaths, possibly related to the systemic toxic effect of the chemotherapeutic agents they were given. Leuko- penia, thrombocytopenia, bleeding, and sepsis developed in these patients. Because patients treated with the usual high-dose chemotherapy regimens were not responding successfully and often had toxic results from this therapy, we began to use low doses of chlorambucil and prednisone. Our early results and those of others 8 confirmed that there was some advantage in treating the S6zary syndrome this way. Our current summary of the patients in this series supports the benefits produced by this treatment regimen. We believe that our improved results with pre-S6zary syn- drome indicate early treatment is desirable. 9 We have not used this regimen to treat patients with T cell leukemia with or without erythroderma.

The current series of results indicates that al- though not all patients had a response to chloram- bucil and prednisone, most did respond, and that duration of life was much longer than that of the patients not treated this way. Contrary to our early clinical impressions, x-ray therapy and chemo- therapy did not provide increased prolonged ben- efit.

Our impression is that this is a safe and useful program capable of producing remissions of S6z- ary syndrome for some patients. Our belief is that, as with all disease, early treatment has advan- tages. Complications include temporary leukope- nia in a few patients. We believe that the occur- rence of sepsis as a cause of death is related more to progressive disease than to treatment. Sepsis occurred equally in our patients treated with or without chlorambucil and prednisone. The possi-

1004 Winkehnann et al

Journal of the American Academy of

Dermatology

bility of t umor format ion in patients who are im- munosuppressed or who are receiving chemother- apy must be r emembered . Because the incidence o f progress ive disease and associated lymphoma was greater in patients not treated wi th low doses o f chlorambuci l and p redn isone , we bel ieve that the low-dose ch lorambuci l -prednisone regimen does not necessar i ly increase such risks in patients with S6zary syndrome.

It is clear the S6zary syndrome and pre-S6zary syndrome respond well to chlorambuci l and pred- nisone in m a ny instances. Reduct ion of Sgzary cell count , improvement o f skin, bo th clinically and his tological ly , and reduct ion o f lymphade- nopathy can be ach ieved by the use o f this low- dose chemotherapeut ic p rogram. Because of the success wi th leukapheresis in some patients, ~~ a p rogram combin ing these two factors might be o f great value. In addi t ion, a program combining t i lorone, d e o x y c o f o r m y c i n , or other a n t i - T cell drugs could be of use. W e have developed a pro- gram that can serve as a basis fo r comparison with other newly developed t rea tment agents and pro- grams. We bel ieve that this program can be the basis for p rospec t ive compar i son studies in the treatment o f S6zary s y n d r o m e that will aid in the deve lopment o f new therapy and help confirm the data presented in this repor t .

REFERENCES

1. Winkelmann RK, Perry HO, Muller SA, et al: Treatment of Sfizary syndrome. Mayo Clin Proc 49:590-592, 1974.

2. Crotty CP, Winkelmann RK: Tilorone hydrochloride in the treatment of T cell lymphoproliferative cutaneous disease. J AM ACAD DERMATOL 7:468-477, 1982.

3. Buechner SA, Winkelmann RK: S6zary syndrome: A clinicopathologic study of 39 cases. Arch Dermatol 119:589-596, 1983.

4. Epstein EH HR, Levin DL, Croft, JD Jr, Lutzner MA: Mycosis fungoides: Survival, prognostic features, re- sponse to therapy, and autopsy findings. Medicine 51:61-72, 1972.

5. Moschella SL: The present status of chemotherapy in dermatology. Med Clin North Am 56:(Pt. 3):725-737, 1972.

6. Edelson RL: Cutaneous T-cell lymphoma. Dermatol Up- date 1: I95-207, 1979.

7. Hoppe RT, Fuks Z, Bagshaw MA: Radiation therapy in the management of cutaneous T-cell lymphomas. Cancer Treat Rep 63:625-632, 1979.

8. Hamminga L, Hartgrink-Groeneveld CA, Van Vloten WA: S6zary's syndrome: A clinical evaluation of eight patients. Br J Dermatol 100:291-296, 1979.

9. Buechner SA, Winkelmann RK: Pre-S6zary erythro- derma evolving to S6zary syndrome. Arch Dermatol 119:285-291, 1983.

10. Winkelmann RK, Revuz J, Pineda AA, Touraine R: Leukapheresis in Sezary syndrome, in Goos M, Chris- tophers E, editors; Lymphoproliferative diseases of the skin. New York, 1982, Springer-Verlag, pp. 228-294.

ABSTRACTS

Dermatit is artefacta in mother and baby as child abuse

Jones DP: Br J Psychiatry 143: 199-200, 1983

Artifactual injuries of the skin were produced in both mother and child as a pathetic form of child abuse. Artifactual dermatoses can be a difficult diagnostic challenge, some- times, for even the most experienced dermatologist.

P.C.A.

Hand eczema in nickel-sensit ive female twins. Genetic predisposit ion and env ironmenta l factors

Menn6 T, Holm NV: Contact Dermatitis 9:289-296, 1983

From the famous Danish Twin Register comes this study of cutaneous allergy to nickel salts. Thirty-four monozygotie probands and forty-five dizygotics were diagnosed. By care-

ful analysis, the authors decided that the number of affected co-twins was not importantly increased over the expectation for the ordinary population. Genetic predisposition does not seem to influence the selection of specific contact antigens.

P.C.A.

Blastomyeosis: A new endemic focus in Canada

Kane J, Righter J, Krajden S, Lester RS: Can Med Assoc J 129:728-731, 1983

North and east of Lake Superior is a principal focus of North American blastomyeosis. Between 1970 and 1981, these thirty-eight cases were collected and treated success- fully about 80% of the time. Delay in diagnosis accounted for some part of the poor prognosis. Specific education, now that this geographic focus is known, can improve therapy.

P.C.A.