II
I
9
|
Extracorporeal photopheresis for the treatment of cutaneous T-cell lymphoma J. Zic, MD, a C. Arzubiaga, MD, a K. E. Salhany, MD, b R. A. Parker, PhD, c D. Wilson, MD, a G. P. Stricklin, MD, PhD, a J. Greer, MD, b and L. E. King, Jr., MD, P h D a Nashville, Tennessee
Background: Many regimens are used for cutaneous T-cell lymphoma (CTCL), but with advanced disease response rates and patient survival are not adequate with any current therapy. Recently extracorporeal photochemotherapy (ECP) was proposed as an alternative therapy. Objective: Our purpose is to present the results of ECP in patients with CTCL refractory to other treatments. Methods: Patients with CTCL received ECP at 3- to 5-week intervals for at least 6 months. All patients except one were in stage T2 (patch/plaque) or higher. Eight patients had extracutaneous disease involving lymph nodes (six patients), bone marrow (five), or S6zary cells (six). The interval between initial symptoms and diagnosis was 5.9 + 1.9 years (mean • standard error of the mean) and the interval between diagnosis and ECP was 2.2 + 0.4 years. Results: A complete response (disappearance of all lesions) was obtained in five patients (25%) and a partial response (disappearance of at least 50% of lesions) in six patients (30%). Four patients (20%) showed stabilization of their disease and five progressed (25%). The only variable that predicted responders versus nonresponders was the number of ECP sessions (/9 < 0.05 by multivariate logistic regression). In contrast, no separate beneficial effect of adjunctive chemotherapy (p > 0.5) or electron beam therapy (p > 0.1) was found. Conclusion: Long-term ECP may be an effective alternative treatment for CTCL refractory to other therapies and is likely to be even more useful when combined with other modalities. (J AM ACAD DERMATOL1992;27:729-36.) In 1975 Edelson et al. 1 proposed that mycosis fungoides (MF), Sgzary syndrome, and related, predominantly cutaneous, T-ceU non-Hodgkin's lymphomas be grouped under the broader classification of cutaneous T-cell lymphoma (CTCL). A recent epidemiologic report by Weinstock and H o r m 2 revealed a 3.2-fold increase in the incidence of a cutaneous form of non-Hodgkin's T-cell lymphoma from 1969 to 1971 to 1984. Unfortunately, data from 1973 to 1984 showed no consistent evidence of improved survival (median survival, 7.0 years) in patients with MF. 3 Treatment for M F and S6zary syndrome includes both topical and systemic therapies. Photochemotherapy (PUVA), 4, 5 topical chemotherapy From the Divisionsof Dermatology,a Oncology and Cell Biology,t' and Biostatistics,e Departments of Medicine, Pathology, and Preventive Medieine, Vanderbilt University and Veterans Administration Medical Center, Nashville. Reprint requests: Lloyd E. King Jr., MD, PhD, 3900 TVC, Divisionof Dermatology, Vantlerbilt University, Nashville, TN 37232.
(nitrogen mustard), 6,7 and electron beam radiation therapy 8' 9 have demonstrated impressive response rates, but most patients treated with topical therapy relapse. 10 Despite advances in combination systemic chemotherapy I 1, 12 and combined modalities (electron beam radiation and chemotherapy), 13, 14 the long-term survival for patients with advanced C T C L (stages II-IV) has remained unchanged. No combination of systemic and topical therapy has been shown to be curative for patients with disseminated disease. 15 Recent evidence has emerged that treatment with interferons alfa-2a and gamma may result in a significant response in patients with CTCL, including many with refractory disease. 16-18 Although toxic side effects of interferons are less serious than other therapies, they are still significant, especially in patients with advanced disease, l~ In photopheresis or extracorporeal photochemotherapy (ECP) leukocyte-enriched peripheral blood is exposed in an extracorporeal system to photoactivated 8-methoxypsoralen (8-MOP).19 In 1987 Edel729
730
Journal of the American Academy of Dermatology
Zic et aL
Table I. Clinical data of CTCL patients treated with photopheresis Age at
~x (yr)/ PL
Race/Sex
Interval Interval from initial from Dx symptom to of CTCL to I)x of CTCL onset of ECP
1 51/W/M
4 yr
4yr
2 74/W/F
2yr
2 yr
3 85/W/F
2yr
2 yr
4 5Vw/M
2yr
2yr
5 77/B/F
2 yr
4 mo
6 47/W/F
6 mo
2 yr
7 52/W/F
7 yr
9 mo
8 29/W/M
1 yr
2 yr
9 80/W/M
15 yr
7 mo
10 47/B/F
1 yr
6 mo
11 61/W/F
9 yr
6 yr
No. of
EC]P
Stage at start of ECP
Prior therapy
EBRT, NM, IIA C/P, T2Nt Mo MTX, Bo PUVA IIA EBRT, NM, C/P, T2NIM0 PUVA, BI topical BCNU IIA EBRT, NM; C/P, IS T2NjMo Bo EBRT, IB PUVA, T2NoMo MTX, Bo pred, dapsone IVB UVA, TS, IS T2N3MI Bi IB pred, TS T2NoM0 Bo IIA PUVA, NM T2N~Mo BI IB Combination chemotx, T2NoMo ABMT, Bo EBRT, TS IVA T4N3Mo Bi IVA NM, CHOP TaN3Mo B1
1 yr
4 yr
13 78/B/F
2 yr
2 yr
NM, C/P, MTX, TS, IS
treatments/ Status
62/Active
Response from onset of ECP
Status
PR
Living
C/P, PUVA 20
PD
Died, sepsis
EBRT, NM, 20 MTX
PD
Died, sepsis
MTX, TS
42/Weaned
PR
Living
TS
31/Weaned
CR
Living
PUVA, NM
28/Weaned
CR
Living
NM, TS
32/Active
PR
Living
NM, 22/Weaned PUVA, IS
CR
Living
EBRT, TS
24/Weaned
CR
Living
18
PD
Died, CTCL
16/Inactive
PD
Living
28*/Active
PR
Living
Combination chemotx, MTX, PUVA, TS IVA EBRT, NM, Combination TaN3Mo PUVA, chemotx, Bo DCF, EBRT,
WN 12 69/W/F
Adjunctive therapy
C/P, NM, TS NM
IIA NM, T2NtMo PUVA, Bo EBRT IB EBRT, C/P, NM T2NoMo NM Bo
20/Weaned, StabRi- Living relapsed zation
ABMT, Autologousbonemarrow transplant; BCNLr, carmustine; CHOP, eyclophosphamide,adriamycin, vincristine, prednisone; Combination ehemotx, combmafio~systemicchemotherapy;C/P, chlorambuciland prednisone;CR, completeresponse;DCF, deoxyeoformycin;DX, diagnosis; EBRT, electronbeamradiationtherapy; ECP, extracorporealphotochemotherapy(photopheresis);IFN, interferon; IS, intralesional steroids;MI, myocardialinrarcfion;MTX, methotrexate;NM, nitrogenmustard;PD, progressivedisease;PR, partial response;pred, prednisone;PUPA, psoralen plus UVAtherapy;,TS, topicalsteroids.
*Patient 12received10ECP treatmentsat MedicalCollegeof Wisconsin.Diseasein patients3, 4, 8, 14,and 16transformedto large-celllymphoma.
Volume 27 Number 5, Part 1 November 1992
Photopheresis in CTCL
731
Table I. Cont'd. Age at
Dx (yr)/ Pt.
Race/Sex
Interval from initial symptom to Dxof~
Interval from Dx of CTCL to onset of ECP
14 6 2 / W / M
10 yr
1 yr
15 5 7 / W / M
2 yr
7 yr
16 4 2 / W / F
4 yr
1 yr
17 67/B/M
i0 yr
1 yr
18 6 2 / W / M
2 mo
6 mo
19 6 6 / W / M
5 yr
8 mo
20 6 5 / W / M
39 yr
5 yr
Stage at start of ECP
IVB T4N3M1 BI IIA T2NIM0 Bo IIA T2NIMo B0 III T4NaMo Bo IVB ToNoMI Bo IIA T2N1Mo Bo IB T2NoMo Bo
Prior therapy
Combination chemotx, EBRT
AdJtmcttve therapy EBRT, TS, chemotx
EBRT, C/P, TS
No. of EC'P treatments/ Status
Response from onset of ECP
12
PD
18/Active CR
Sta~s
Died, MI Living
NM Herbals, vitamins
PUVA
EBRT, C/P, c/P NM, TS
22/Active Stabili- Living zation 22/Active Stabili- Living zation
NM
TS
14/Active Stabili- Living zation
EBRT, NM, TS
TS, C/P
16/Active PR
Living
EBRT, TS, PUVA, NM
NM
14/Active PR
Living
son et al. 2~ reported the first clinical trial of this modality in the treatment of CTCL. Twenty-seven of 37 patients with otherwise resistant CTCL responded. Patients with S6zary syndrome demonstrated the most significant response and improved survival. The effectiveness of ECP for CTCL patients with generalized patch/plaque and tumor stage has not been extensively evaluated. Armus et al. 21 recently published data from eight EC P-t reated patients. Four of five patients with erythrodermic C T C L and aU three patients with patch/plaque or tumor stage responded. We review our experience with E C P in 24 C T C L patients with refractory patch/plaque, tumor, and erythrodermic stage CTCL.
excluded had less than 6 months of therapy at the time of this evaluation. One patient withdrew for personal reasons and one patient could not tolerate repeated venipunctures. Patient 8 had Ki-l-positive anaplastic largecell lymphoma and had failed to respond to radiation therapy and aggressive chemotherapy, including autologous bone marrow transplant. Patients were not excluded on the basis of stage of disease or extent of prior therapy. Each patient had a complete history and physical examination with detailed documentation of skin involvement. Only patients who were free of known viral infections (hepatitis, human T-cell leukemia/lymphoma virus type I, human immunodeficiency virus) were admitted to the Renal Dialysis Center, which housed the ECP machines and shared nursing services.
Morphologic and immunologic studies MATERIAL AND METHODS Patient population Twenty-four patients with CTCL entered the ECP program between September 1988 and April 1991. All patients had CTCL documented by skin biopsy results. Most patients were judged to have disease refractory to standard therapy at the time of referral and demonstrated progressive disease before ECP. This series includes only those 20 patients who had undergone at least 6 months of ECP treatments. Their characteristics and previous therapies are shown in Table I. Two patients who were
Formalin- or B-5-fixed biopsy specimens of skin, lymph nodes, and bone marrow were paraffin embedded and cut in 4 ~zmsections that were stained with hematoxy/in and eosin, periodic acid-Schiff, and methyl green pyronine stains. One of us (K. E. S.) reviewed all pathologic material in each case to provide a uniform interpretation of each patient's disease course. Biopsy specimens from skin, lymph nodes, and bone marrow and Wrightstained peripheral blood smears were also reviewed to identify the sites of involvement and to determine whether any cases had undergone histologic progression or trans-
Journal of the 732
American Academy of Dermatology
Z i c et al.
Table IIA. CTCL Cooperative Group Staging
Table IIB. T N M staging classification stage
T:
Skin To T~
T2
T3
T4 N: Lymph nodes No
Nl
N2
N~
B:
Clinically and/or histopathologically suspect lesions Limited plaques, papules, or eczematous patches covering __10%of skin surface Tumors >" 1 Generalized erythroderma No clinically abnormal peripheral lymph nodes, pathology negative for CTCL Clinically abnormal peripheral lymph nodes, pathology negative for CTCL No clinically abnormal peripheral lymph nodes, pathology positive for CTCL Clinically abnormal peripheral lymph nodes, pathology positive for CTCL
Peripheral blood
B0 Bl M: Visceralorgans M0 Mt
Atypical circulating cells not present (5%) No visceral organ involvement Visceral involvement (must have pathology confirmation and organ involved should be specified
Data from Ramsay DL, HalperinPS, Zeleniuch-JacquotteA. J AM ACADDEaMATOL1988;19:684-91.
formation to the large-cell variant of CTCL, as previously reported. 22
Immunophenotypic analysis was performed on 18 of 20 cases by frozen section immunoperoxidase23or paraffin immunoperoxidase methods.24 The panel used ineluded antibodies to T-cell-associated antigens CD2, CD3, CD4, CD5, CD7, CD8 (Becton-Dicldnson) and CD45RO (UCHLI, Dako). B-cell-associated antigens CD19, CD20 (Coulter), CD22 (Becton-Dickinson), and
.
IA IB IIA IIB III IVA IVB
T I 2 1,2 3 4 0-4 O-4
I
N 0 0 1 0,i 0,1 2,3 O-3
I
M 0 0 0
0 0 0 1
surface immunoglobulins IgG, IgA, IgM, Kand X (Tago); leukocyte common antigen CD45 (Dako); and lymphocyte activation-associated antigens CD 15 (Leu-MI, Becton-Dickinson), CD30 (Ki-l, Dako), and HLA-DR (Becton-Dickinson). Immunophenotypic analysis was done as previously described. 22
Laboratory and staging studies All patients had baseline laboratory studies including complete blood cell count (CBC) with differential and platelet count, electrolytes, liver function tests, as well as determination of levels of blood urea nitrogen, serum creatinine, calcium, and phosphorus. These studies were repeated before each ECP session. Serum protein electrophoresis and quantitative immunoglobulin levels were also obtained. Nineteen of the 20 patients had bone marrow examinations along with abdominal computed tomographic (CT) scans for initial staging. Seven patients had lymph node biopsies. S~zary cell counts were obtained in al/patients and were available in 19 patients. The CTCL Cooperative Group Staging classification (Table II) was used to describe the extent of disease at the onset of ECP. 25 Patients with clinically abnormal lymph nodes and no record of lymph node biopsy were classified as having stage N l disease. No biopsies of clinically normal lymph nodes were performed.
Photopheresis The UVAR II system developed by Therakos (Johnson & Johnson Co., King of Prussia, Pa.) was used. Patients ingested approximately 0.6 mg/kg of 8-MOP 1.5 to 2.0 hours before the initiation of the leukocyte enrichment phase. The patient's blood was collected via a 16-gauge antecubital catheter and separated by continuous centrifugation in six cycles to obtain 240 or 270 ml of leukocyte-enriched blood and 300 ml of plasma. In some patients nine cycles were required to obtain an adequate volume of leukocytes. The nine-cycle setting is a built-in feature of the UVAR II because larger patients may require more leukocyte exchanges. The leukocyte-enriched fraction was combined with 300 ml of plasma and 200 ml of sterile normal saline. The total volume was exposed to a UVA light source for a minimum of 1.5 hours while be-
Volume 27 Number 5, Part 1 November 1992
ing passed as a 1 mm film through a six-chambered, disposable, sterile cassette constructed of acrylic. After exposure of the leukocyte-enriched fraction to UVA light, the entire volume was returned to the patient. All patients had a CBC with differential performed on the leukocyte-enriched fraction exposed to UVA. All patients had serum psoralen levels evaluated after ingestion of their first dose of 8-MOP and at various points in their treatment. Psoralen levels were also measured on the total collected volume of the leukocyte-enriched fraction of six patients (F. Gasparro, PhD, Yale University). The ECP procedure was performed on two consecutive days at 3- to 5-week intervals depending on the severity of the disease. Patients were examined for evidence of clinical response at each ECP session. A complete response was defined as the disappearance of all measurable lesions for at least 1 month. A partial response was defined as greater than 50% disappearance of all measurable lesions for at least 1 month. Stabilization was defined as less improvement of lesions than a partial response but no evidence of new lesions. Progressive disease was defined as the appearance of new skin lesions.
Adjunctive therapy Adjunctive therapy was considered on an individual basis depending on the severity of the disease and initial response to ECP. Adjunctive therapies were not used for the initial 4 to 6 months of the ECP treatment period. Adjunctive therapies included topical nitrogen mustard (10 mg/dl concentration in ointment), oral chlorambucil and prednisone, methotrexate, electron beam radiotherapy, PUVA, combination chemotherapy, and topical/intralesional steroids. The response to the agents used adjunctively was compared before and after ECP to judge additive or synergistic effects.
Statistical analysis To determine whether the clinical, morphologic, immunologic, or treatment variables would predict the final outcome, these were tested against the final outcome defined as either responders (complete or partial remission) or nonresponders (stabilization or progressive disease [see later]). Continuous variables were tested with a t test. Discrete variables were assessed with a chi-square test, or, for two-by-two tables, with Fisher's exact test. Criterion for significance was p < 0.05. Data are presented as mean + standard error of the mean. RESULTS
Patient population Table I summarizes the clinical data for the C T C L patients. The mean age at diagnosis was 61.2 __+ 3.2 years. There were 10 men and 10 women. The mean interval from initial symptom to
Photopheresis in CTCL 733 Table Ill. Morphologic and inamunologic studies Characteristic
No.
Extracutaneous disease Lymph nodes Bone marrow Peripheral blood Transformed CTCL T-cell immunophenotype Helper T-cell subtype Aberrant T-cell phenotype Expression of activation markers
8/20 4/6 5/19 6/19 5/20 18/ 18 16/18 7/18 15/16
diagnosis of CTCL was 5.9 _ 2.0 years. The mean interval from diagnosis of C T C L to onset of E C P was 2.2 +__0.4 years. Three patients (Nos. 8, 19, 20) showed improvement of their clinical staging with therapies before ECP, but were not free of disease. Patients 8, t9, and 20 were described as having clinical stages IVB, III, and IIA, respectively, at the time of diagnosis of CTCL. All other patients were considered to be at their worst clinical stage at the onset of ECP. At the start of ECP 14 patients were in the generalized patch/plaque stage (T2), two were in the tumor stage (T3), and three patients had exfoliative erythroderma (T4). Nix months before ECP, patient 18 had generalized patch/plaquestage disease and biopsy-confirmed bone marrow involvement with CTCL. After 6 months of topical nitrogen mustard therapy the patient's skin lesions improved dramatically (To). ECP was initiated at this stage because of persistent bone marrow involvement at the request of the referring oncologist because no other therapy had been effective and because of the previous success with his patient (8) who had marrow involvement.
Morphologic and immunologic studies The morphologic and immunologic characteristics are summarized in Table III. Eight patients had proven extracutaneous disease with involvement of the lymph nodes (six cases), bone marrow (five cases), and circulating Srzary cells more than 15% of lymphocytes (six cases). In five cases there was evidence of transformation from C T C L to the largecell lymphoma variant. 22 In t8 patients who had complete phenotyping performed, there were 16 T-cell helper subtypes and 7 aberrant T-cell subtypes. In 15 of 16 patients there was expression of T-cell activation markers. 26
Journal of the American Academy of Dermatology
734 Zic et aL Clinical response Eleven patients showed a partial or complete response to ECP. The disease stabilized in four patients and progressed in five patients. Ten of 14 patients with generalized patch/plaque stage responded. Neither of the two patients with tumorstage disease improved. One of the three patients with refractory exfoliative erythroderma responded. Patients were categorized as responders (partial or complete response) and nonresponders (progressive disease or stabiliTation) for the purpose of analysis. The mean age of the responding group was 59.3 + 3.2 years compared with 63.8 +__3.3 years for the nonresponding group (nonsignificant difference). The CTCL patients who responded did so usually within the first 6 months of therapy. Six patients (Nos. 4, 5, 6, 8, 9, and 13) were successfully weaned from ECP. One patient relapsed (No. 13) requiring reinitiation of ECP treatments 4 months after weaning. At this time, patients 4, 5, 6, 8, and 9 remain in clinical remission off ECP treatments for 4, 9, 9, 9, and 3 months, respectively. Patient 4 continues to receive methotrexate therapy. Patient 11 was weaned from ECP after only modest progressive improvement after 16 sessions. However, this patient had a marked progression of disease since weaning from ECP. Comparison of clinical data between responders and nonresponders showed no statistically significant differences between the two groups with regard to race, age, the interval between initial symptom and diagnosis of CTCL, the interval from diagnosis of CTCL to initiation of ECP, the stage of disease, the presence of extracutaneous involvement, previous therapies including electron beam therapy and chemotherapy, the average leukocyte fraction exposed to UVA light, or the mean psoralen serum levels achieved. Mean serum psoralen levels were 153 + 20 mg/ml (range 49 to 285 mg/ml). When analyzed by simple t test, the only variable significantly associated with outcome (response vs no response) was the number of ECP sessions (t9 = 0.025). To determine whether this difference was associated with other variables, logistic regression was also performed. With this approach, the number of ECP sessions was again the variable that best predicted outcome (.p < 0.05). In contrast, no additional or separate beneficial effect of adjunctive
chemotherapy (p > 0.5) or preceding electron beam therapy (p > 0.1) was found. Adverse effects of ECP More than 500 ECP treatments were performed with minimal patient complications or adverse effects. The most common side effect was nausea after ingestion of 8-MOP. The nausea was transient and rarely interfered with treatment. Other less common comphcations included hypotension responsive to fluid administration (three patients), hypoglycemia (two patients), and hematomas at catheter sites (three patients). One patient (No. 18) experienced hemolysis during an ECP session resulting in hemoglobinuria, but without a decline in renal function. An extensive evaluation revealed no identifiable cause for the episode other than mechanically induced hemolysis confirmed by Therakos after examination of the photocassette. This patient tolerated subsequent ECP treatments without complications. Adjunctive therapy ECP improved or stabilized the clinical status of 15 of 20 patients if judged by previous therapy status. In combination with ECP 10 of 20 patients were able to be maintained with minimal additional therapy: topical steroids alone (patients 5, 15, and 18); PUVA alone (patient 16); nitrogen mustard alone (patients 12, 13, and 20); topical steroids and nitrogen mustard (patient 7); topical steroids and methotrexate (patient 4); and topical steroids combined with chlorambucil and prednisone (patient 19). DISCUSSION The beneficial effect of ECP for CTCL patients with S6zary syndrome has been clearly established.20, 21, 26-29FoUow-up studies by Heald et al. 26 found that erythrodermic patients did better when ECP was started earlier in the course of the disease and before patients were heavily pretreated. Only one of the three heavily treated patients with erythrodermic CTCL in our series responded to ECP and required adjunctive therapy. The beneficial effect of ECP for established tumor-stage CTCL patients has been less encouraging. However, some patients reportedly have had favorable responses when EQP was combined with electron beam radiotherapy.21, 26 Both patients in
Volume 27 Number 5, Part 1 November 1992
this series with advanced tumor-stage disease who already had other conventional and experimental treatments showed no response to ECP even with adjunctive combination chemotherapy, PUVA, and external beam radiotherapy. W e report the largest series thus far of CTCL patients with generalized patch/plaque stage (T2) disease treated with ECP. Of the 14 patients in this series with generalized patch/plaque-stage disease, 10 had a favorable response to ECP. Four patients demonstrated a complete response and six had a partial response. Four of the 10 responding patients have been successfully weaned from ECP and remain in remission at this time. Most patients in this series received extensive therapy before ECP and all had either shown no response or had progression of disease during these therapies. It was unclear from the data whether certain previous therapies made CTCL patients more sensitive to the beneficial effects of ECP. Establishing the degree to which adjunctive therapy will affect the outcome of the patients in this series is also difficult. However, 10 of 20 patients were able to be maintained with only minimal adjunctive therapy, whereas they had unstable disease before ECP. Patients received different adjunctive therapies for a variable time after ECP in this series. Adjunctive therapy with ECP has not been systematically evaluated in the treatment of CTCL. Heald et al. 26 suggested that low-dose methotrexate may be the best adjunctive therapy with ECP because it is relatively nontoxic and nonimmunosuppressive. The patients in this series receiving methotrexate did not demonstrate any striking difference in response compared with those patients not receiving the drug. In our series, the only consistent therapeutic intervention during the period of observation in this series was ECP. Therefore we attributed a major portion of the patients' responses to this treatment modality but believe that adjunctive therapy is helpful. Multivariate analysis revealed that the only factor that correlated significantly with the final outcome was the number of ECP sessions the patients received. Because we did not systematically evaluate the effect of adjunctive therapy, we cannot precisely define the contribution of any single agent other than ECP to the final outcome. W e observed that four of five CTCL patients with stage IB disease (generalized patch/plaque, no elin-
Photopheresis in CTCL 735 icaUy abnormal lymph nodes) responded to ECP. This suggests that ECP may be important in the treatment of this subgroup of patients with disease apparently localized to the skin. We also observed that a patient with Ki-l-positive lymphoma is free of disease after failing to respond to radiation and bone marrow transplantation. Our experience with ECP supports the conclusion of others 19-21,26"29 that ECP is a relatively safe modality with infrequent complications. Complications associated with other treatment modalities for CTCL such as myelosuppression, fiver toxicity, renal toxicity, neural toxicity, and radiation dermatitis have not been reported with ECP. The hemolytic episode that occurred during an ECP session was probably due to mechanical hemolysis and not from ECP itself. The mechanism of the response to ECP has not been completely elucidated. Edelson noted that it appears to involve an induced immunologic reaction against the patient's malignant T cells. 19,28 Photoactivated 8-MOP forms covalent pfiotoadducts with nuclear DNA, thereby affecting the antigenicity and immunogenicity of exposed nucleated cells.19 Evidence from animal and human studies suggests that the reinfusion of the irradiated T cells induces a therapeutically significant immunologic reaction that targets the malignant clones. 19,28,29 Our studies did not address these mechanisms. REFERENCES
1. Edelson RL. CTCL: the Srzary syndrome, mycosis fungoides, and related disorders. Ann Intern Med 1975;83:53452. 2. Weinstock MA, Horrn JW. Mycosis fungoides in the United States: increasing incidence and descriptive epidemiology. JAMA 1988;260:42-6. 3. Weinstock MA, Horm JW. Population-based estimate of survival and determinants of prognosis in patients with mycosis fungoides. Cancer 1988;62:1658-61. 4. Rosenbaum MM, Roenigk H H Jr, Caro WA, et al. Photochemotherapy in cutaneous T cell lymphoma and parapsoriasis en plaques. J AM ACAD DERMATOL1985;13:61322. 5. Hrnigsmann H, Brenner W, Rauschmeier W, et al. Photochemotherapy for cutaneous T cell lymphoma. J AM ACAD DERMATOL 1984; 10:238-45. 6. RamsayDL, HalperinPS, Zeleniuch-JacquotteA. Topical mechlorethamine therapy for early stage mycosis fungoides. J AM ACAD DERMATOL 1988;19:684-91. 7. Vonderheid EC, Tan ET, Kantor AF, et al. Long-term efficacy, curative potential, and carcinogenicity of topical mechlorethamine chemotherapy in cutaneous T cell lymphoma. J AM ACAD DERMATOL 1989;20:416-28. 8. Cotter GW, Baglan R J, Wasserman TH, et al. PaUiative
736
9. 10. 11. 12. 13.
14. 15. 16. 17.
18.
Z i c et al.
radiation treatment of cutaneous mycosis fungoides: a dose response. Int J Radiat Oncol Bid Phys 1983;9:1477-80. Hoppe RT, Cox TS, Fuks Z, et al. Electron-beam therapy for mycosis fungoides: the Stanford University experience. Cancer Treat Rep 1979;63:691-700. Kuzel TM, Springer E, Roenigk HH Jr, et al. Mycosis fungoides: time to define the best therapy. J Natl Cancer Inst 1990;82:183-5. Grozea PN, Jones SE, MeKelvey EM, et al. Combination chemotherapy for mycosis fungoides: a Southwest Oneology Group study. Cancer Treat Rep 1979;61:279-82. Minna JD, Roenigk HH Jr, Glastein E. Report of the Committee of Therapy for Mycosis Fungoidez and S~zary Syndrome. Cancer Treat Rep 1979;63:729-36. Bunn PA Jr, Fischmann AB, Schechter GP, et al. Combined modality therapy with electron-beam irradiation and systemic chemotherapy for cutaneous T cell lymphoma. Cancer Treat Rep 1979;63:713-7. Braverman IM, Yager NB, Chen M, eta[. Combined total body electron beam irradiation and chemotherapy for mycosis fungoides. J AM ACAD DERMATOL1987;16:45-60. Knobler RM, Edelson RL. Cutaneous T cell lymphoma. Med Clin North Am 1986;70:109-38. OIsen EA, Rosen ST, VoUmer RT, et al. Interferon alfa-2a in the treatment of cutaneous T cell lymphoma. J AM ACAD DERMATOL 1989;20:395-407. Kaplan EH, Rosen ST, Norris DB, et al. Phase I[ study of recombinant human interferon gamma for treatment of cutaneous T cell lymphoma. J Natl Cancer Inst 1990; 82:208-12. Kuzel TM, Gilyon K, Springer E, et al. Interferon alfa-2a combined with phototherapy in the treatment of cutaneous T ceil lymphoma. J Natl Cancer Inst 1990;82:203-7.
Journal of the American Academy of Dermatology
19. Edelson RL. Photopheresis: a new therapeutic concept. Yale J Biol Med 1989;62:565-77. 20. Edelson R, Berger C, Gasparro F, et al. Treatment of cutaneous T cell lymphoma by extracorpereal photochemotherapy. N Engl J Med 1987;316:297-303. 21. Armus S, Keyes B, Cahill C, et al. Photopheresis for the treatment of cutaneous T cell lymphoma. J AM ACAD DERMATOL1990;23:898-902. 22. Salhany KE, Cousar JB, Greer JP, et al. Transformation of cutaneous T cell lymphoma to large ceU lymphoma: a clinicopathologic and immunologic study. Am J Pathol 1988; 132:265-77. 23. Borowitz M J, Croker BP, Burchette J. Immunocytochernicaldetection of lymphocytesurface antigens in fixed tissue sections. J Histochem Cytochem 1982;30:171-4. 24. Pinkus GS, Thomas P, Said JW. Leu-Ml--a marker for Reed-Sternberg cells in Hodgkin's disease: an immunoperoxidase study of paraffin-embedded tissues. Am J Pathol 1985;I 19:244-52. 25. Bunn PA, Lamberg SI. Report of the committee on staging and classification of cutaneous T cell lymphomas. Cancer Treat Rep 1979;63:725-7. 26. Heald PW, Perez MI, Christensen I, et al. Photopheresis therapy of cutaneous T cell lymphoma: the Yale-New Haven Hospital experience. Yale J Biol Med 1989;62:629-38. 27. Knobler RM. Photopheresis--extracorporeal irradiation of 8-MOP containing blood: a new therapeutic modality. Blut 1987;54:247-50. 28. Heald PW, Edelson RL. Photopheresis therapy of cutaneous T cell lymphoma. Keio J Med 1988;37:155-67. 29. Edelson RL. Photophercsis. J Clin Apheresis 1990;5:77-9.
I
9
|
Extracorporeal photopheresis for the treatment of cutaneous T-cell lymphoma J. Zic, MD, a C. Arzubiaga, MD, a K. E. Salhany, MD, b R. A. Parker, PhD, c D. Wilson, MD, a G. P. Stricklin, MD, PhD, a J. Greer, MD, b and L. E. King, Jr., MD, P h D a Nashville, Tennessee
Background: Many regimens are used for cutaneous T-cell lymphoma (CTCL), but with advanced disease response rates and patient survival are not adequate with any current therapy. Recently extracorporeal photochemotherapy (ECP) was proposed as an alternative therapy. Objective: Our purpose is to present the results of ECP in patients with CTCL refractory to other treatments. Methods: Patients with CTCL received ECP at 3- to 5-week intervals for at least 6 months. All patients except one were in stage T2 (patch/plaque) or higher. Eight patients had extracutaneous disease involving lymph nodes (six patients), bone marrow (five), or S6zary cells (six). The interval between initial symptoms and diagnosis was 5.9 + 1.9 years (mean • standard error of the mean) and the interval between diagnosis and ECP was 2.2 + 0.4 years. Results: A complete response (disappearance of all lesions) was obtained in five patients (25%) and a partial response (disappearance of at least 50% of lesions) in six patients (30%). Four patients (20%) showed stabilization of their disease and five progressed (25%). The only variable that predicted responders versus nonresponders was the number of ECP sessions (/9 < 0.05 by multivariate logistic regression). In contrast, no separate beneficial effect of adjunctive chemotherapy (p > 0.5) or electron beam therapy (p > 0.1) was found. Conclusion: Long-term ECP may be an effective alternative treatment for CTCL refractory to other therapies and is likely to be even more useful when combined with other modalities. (J AM ACAD DERMATOL1992;27:729-36.) In 1975 Edelson et al. 1 proposed that mycosis fungoides (MF), Sgzary syndrome, and related, predominantly cutaneous, T-ceU non-Hodgkin's lymphomas be grouped under the broader classification of cutaneous T-cell lymphoma (CTCL). A recent epidemiologic report by Weinstock and H o r m 2 revealed a 3.2-fold increase in the incidence of a cutaneous form of non-Hodgkin's T-cell lymphoma from 1969 to 1971 to 1984. Unfortunately, data from 1973 to 1984 showed no consistent evidence of improved survival (median survival, 7.0 years) in patients with MF. 3 Treatment for M F and S6zary syndrome includes both topical and systemic therapies. Photochemotherapy (PUVA), 4, 5 topical chemotherapy From the Divisionsof Dermatology,a Oncology and Cell Biology,t' and Biostatistics,e Departments of Medicine, Pathology, and Preventive Medieine, Vanderbilt University and Veterans Administration Medical Center, Nashville. Reprint requests: Lloyd E. King Jr., MD, PhD, 3900 TVC, Divisionof Dermatology, Vantlerbilt University, Nashville, TN 37232.
(nitrogen mustard), 6,7 and electron beam radiation therapy 8' 9 have demonstrated impressive response rates, but most patients treated with topical therapy relapse. 10 Despite advances in combination systemic chemotherapy I 1, 12 and combined modalities (electron beam radiation and chemotherapy), 13, 14 the long-term survival for patients with advanced C T C L (stages II-IV) has remained unchanged. No combination of systemic and topical therapy has been shown to be curative for patients with disseminated disease. 15 Recent evidence has emerged that treatment with interferons alfa-2a and gamma may result in a significant response in patients with CTCL, including many with refractory disease. 16-18 Although toxic side effects of interferons are less serious than other therapies, they are still significant, especially in patients with advanced disease, l~ In photopheresis or extracorporeal photochemotherapy (ECP) leukocyte-enriched peripheral blood is exposed in an extracorporeal system to photoactivated 8-methoxypsoralen (8-MOP).19 In 1987 Edel729
730
Journal of the American Academy of Dermatology
Zic et aL
Table I. Clinical data of CTCL patients treated with photopheresis Age at
~x (yr)/ PL
Race/Sex
Interval Interval from initial from Dx symptom to of CTCL to I)x of CTCL onset of ECP
1 51/W/M
4 yr
4yr
2 74/W/F
2yr
2 yr
3 85/W/F
2yr
2 yr
4 5Vw/M
2yr
2yr
5 77/B/F
2 yr
4 mo
6 47/W/F
6 mo
2 yr
7 52/W/F
7 yr
9 mo
8 29/W/M
1 yr
2 yr
9 80/W/M
15 yr
7 mo
10 47/B/F
1 yr
6 mo
11 61/W/F
9 yr
6 yr
No. of
EC]P
Stage at start of ECP
Prior therapy
EBRT, NM, IIA C/P, T2Nt Mo MTX, Bo PUVA IIA EBRT, NM, C/P, T2NIM0 PUVA, BI topical BCNU IIA EBRT, NM; C/P, IS T2NjMo Bo EBRT, IB PUVA, T2NoMo MTX, Bo pred, dapsone IVB UVA, TS, IS T2N3MI Bi IB pred, TS T2NoM0 Bo IIA PUVA, NM T2N~Mo BI IB Combination chemotx, T2NoMo ABMT, Bo EBRT, TS IVA T4N3Mo Bi IVA NM, CHOP TaN3Mo B1
1 yr
4 yr
13 78/B/F
2 yr
2 yr
NM, C/P, MTX, TS, IS
treatments/ Status
62/Active
Response from onset of ECP
Status
PR
Living
C/P, PUVA 20
PD
Died, sepsis
EBRT, NM, 20 MTX
PD
Died, sepsis
MTX, TS
42/Weaned
PR
Living
TS
31/Weaned
CR
Living
PUVA, NM
28/Weaned
CR
Living
NM, TS
32/Active
PR
Living
NM, 22/Weaned PUVA, IS
CR
Living
EBRT, TS
24/Weaned
CR
Living
18
PD
Died, CTCL
16/Inactive
PD
Living
28*/Active
PR
Living
Combination chemotx, MTX, PUVA, TS IVA EBRT, NM, Combination TaN3Mo PUVA, chemotx, Bo DCF, EBRT,
WN 12 69/W/F
Adjunctive therapy
C/P, NM, TS NM
IIA NM, T2NtMo PUVA, Bo EBRT IB EBRT, C/P, NM T2NoMo NM Bo
20/Weaned, StabRi- Living relapsed zation
ABMT, Autologousbonemarrow transplant; BCNLr, carmustine; CHOP, eyclophosphamide,adriamycin, vincristine, prednisone; Combination ehemotx, combmafio~systemicchemotherapy;C/P, chlorambuciland prednisone;CR, completeresponse;DCF, deoxyeoformycin;DX, diagnosis; EBRT, electronbeamradiationtherapy; ECP, extracorporealphotochemotherapy(photopheresis);IFN, interferon; IS, intralesional steroids;MI, myocardialinrarcfion;MTX, methotrexate;NM, nitrogenmustard;PD, progressivedisease;PR, partial response;pred, prednisone;PUPA, psoralen plus UVAtherapy;,TS, topicalsteroids.
*Patient 12received10ECP treatmentsat MedicalCollegeof Wisconsin.Diseasein patients3, 4, 8, 14,and 16transformedto large-celllymphoma.
Volume 27 Number 5, Part 1 November 1992
Photopheresis in CTCL
731
Table I. Cont'd. Age at
Dx (yr)/ Pt.
Race/Sex
Interval from initial symptom to Dxof~
Interval from Dx of CTCL to onset of ECP
14 6 2 / W / M
10 yr
1 yr
15 5 7 / W / M
2 yr
7 yr
16 4 2 / W / F
4 yr
1 yr
17 67/B/M
i0 yr
1 yr
18 6 2 / W / M
2 mo
6 mo
19 6 6 / W / M
5 yr
8 mo
20 6 5 / W / M
39 yr
5 yr
Stage at start of ECP
IVB T4N3M1 BI IIA T2NIM0 Bo IIA T2NIMo B0 III T4NaMo Bo IVB ToNoMI Bo IIA T2N1Mo Bo IB T2NoMo Bo
Prior therapy
Combination chemotx, EBRT
AdJtmcttve therapy EBRT, TS, chemotx
EBRT, C/P, TS
No. of EC'P treatments/ Status
Response from onset of ECP
12
PD
18/Active CR
Sta~s
Died, MI Living
NM Herbals, vitamins
PUVA
EBRT, C/P, c/P NM, TS
22/Active Stabili- Living zation 22/Active Stabili- Living zation
NM
TS
14/Active Stabili- Living zation
EBRT, NM, TS
TS, C/P
16/Active PR
Living
EBRT, TS, PUVA, NM
NM
14/Active PR
Living
son et al. 2~ reported the first clinical trial of this modality in the treatment of CTCL. Twenty-seven of 37 patients with otherwise resistant CTCL responded. Patients with S6zary syndrome demonstrated the most significant response and improved survival. The effectiveness of ECP for CTCL patients with generalized patch/plaque and tumor stage has not been extensively evaluated. Armus et al. 21 recently published data from eight EC P-t reated patients. Four of five patients with erythrodermic C T C L and aU three patients with patch/plaque or tumor stage responded. We review our experience with E C P in 24 C T C L patients with refractory patch/plaque, tumor, and erythrodermic stage CTCL.
excluded had less than 6 months of therapy at the time of this evaluation. One patient withdrew for personal reasons and one patient could not tolerate repeated venipunctures. Patient 8 had Ki-l-positive anaplastic largecell lymphoma and had failed to respond to radiation therapy and aggressive chemotherapy, including autologous bone marrow transplant. Patients were not excluded on the basis of stage of disease or extent of prior therapy. Each patient had a complete history and physical examination with detailed documentation of skin involvement. Only patients who were free of known viral infections (hepatitis, human T-cell leukemia/lymphoma virus type I, human immunodeficiency virus) were admitted to the Renal Dialysis Center, which housed the ECP machines and shared nursing services.
Morphologic and immunologic studies MATERIAL AND METHODS Patient population Twenty-four patients with CTCL entered the ECP program between September 1988 and April 1991. All patients had CTCL documented by skin biopsy results. Most patients were judged to have disease refractory to standard therapy at the time of referral and demonstrated progressive disease before ECP. This series includes only those 20 patients who had undergone at least 6 months of ECP treatments. Their characteristics and previous therapies are shown in Table I. Two patients who were
Formalin- or B-5-fixed biopsy specimens of skin, lymph nodes, and bone marrow were paraffin embedded and cut in 4 ~zmsections that were stained with hematoxy/in and eosin, periodic acid-Schiff, and methyl green pyronine stains. One of us (K. E. S.) reviewed all pathologic material in each case to provide a uniform interpretation of each patient's disease course. Biopsy specimens from skin, lymph nodes, and bone marrow and Wrightstained peripheral blood smears were also reviewed to identify the sites of involvement and to determine whether any cases had undergone histologic progression or trans-
Journal of the 732
American Academy of Dermatology
Z i c et al.
Table IIA. CTCL Cooperative Group Staging
Table IIB. T N M staging classification stage
T:
Skin To T~
T2
T3
T4 N: Lymph nodes No
Nl
N2
N~
B:
Clinically and/or histopathologically suspect lesions Limited plaques, papules, or eczematous patches covering __10%of skin surface Tumors >" 1 Generalized erythroderma No clinically abnormal peripheral lymph nodes, pathology negative for CTCL Clinically abnormal peripheral lymph nodes, pathology negative for CTCL No clinically abnormal peripheral lymph nodes, pathology positive for CTCL Clinically abnormal peripheral lymph nodes, pathology positive for CTCL
Peripheral blood
B0 Bl M: Visceralorgans M0 Mt
Atypical circulating cells not present (5%) No visceral organ involvement Visceral involvement (must have pathology confirmation and organ involved should be specified
Data from Ramsay DL, HalperinPS, Zeleniuch-JacquotteA. J AM ACADDEaMATOL1988;19:684-91.
formation to the large-cell variant of CTCL, as previously reported. 22
Immunophenotypic analysis was performed on 18 of 20 cases by frozen section immunoperoxidase23or paraffin immunoperoxidase methods.24 The panel used ineluded antibodies to T-cell-associated antigens CD2, CD3, CD4, CD5, CD7, CD8 (Becton-Dicldnson) and CD45RO (UCHLI, Dako). B-cell-associated antigens CD19, CD20 (Coulter), CD22 (Becton-Dickinson), and
.
IA IB IIA IIB III IVA IVB
T I 2 1,2 3 4 0-4 O-4
I
N 0 0 1 0,i 0,1 2,3 O-3
I
M 0 0 0
0 0 0 1
surface immunoglobulins IgG, IgA, IgM, Kand X (Tago); leukocyte common antigen CD45 (Dako); and lymphocyte activation-associated antigens CD 15 (Leu-MI, Becton-Dickinson), CD30 (Ki-l, Dako), and HLA-DR (Becton-Dickinson). Immunophenotypic analysis was done as previously described. 22
Laboratory and staging studies All patients had baseline laboratory studies including complete blood cell count (CBC) with differential and platelet count, electrolytes, liver function tests, as well as determination of levels of blood urea nitrogen, serum creatinine, calcium, and phosphorus. These studies were repeated before each ECP session. Serum protein electrophoresis and quantitative immunoglobulin levels were also obtained. Nineteen of the 20 patients had bone marrow examinations along with abdominal computed tomographic (CT) scans for initial staging. Seven patients had lymph node biopsies. S~zary cell counts were obtained in al/patients and were available in 19 patients. The CTCL Cooperative Group Staging classification (Table II) was used to describe the extent of disease at the onset of ECP. 25 Patients with clinically abnormal lymph nodes and no record of lymph node biopsy were classified as having stage N l disease. No biopsies of clinically normal lymph nodes were performed.
Photopheresis The UVAR II system developed by Therakos (Johnson & Johnson Co., King of Prussia, Pa.) was used. Patients ingested approximately 0.6 mg/kg of 8-MOP 1.5 to 2.0 hours before the initiation of the leukocyte enrichment phase. The patient's blood was collected via a 16-gauge antecubital catheter and separated by continuous centrifugation in six cycles to obtain 240 or 270 ml of leukocyte-enriched blood and 300 ml of plasma. In some patients nine cycles were required to obtain an adequate volume of leukocytes. The nine-cycle setting is a built-in feature of the UVAR II because larger patients may require more leukocyte exchanges. The leukocyte-enriched fraction was combined with 300 ml of plasma and 200 ml of sterile normal saline. The total volume was exposed to a UVA light source for a minimum of 1.5 hours while be-
Volume 27 Number 5, Part 1 November 1992
ing passed as a 1 mm film through a six-chambered, disposable, sterile cassette constructed of acrylic. After exposure of the leukocyte-enriched fraction to UVA light, the entire volume was returned to the patient. All patients had a CBC with differential performed on the leukocyte-enriched fraction exposed to UVA. All patients had serum psoralen levels evaluated after ingestion of their first dose of 8-MOP and at various points in their treatment. Psoralen levels were also measured on the total collected volume of the leukocyte-enriched fraction of six patients (F. Gasparro, PhD, Yale University). The ECP procedure was performed on two consecutive days at 3- to 5-week intervals depending on the severity of the disease. Patients were examined for evidence of clinical response at each ECP session. A complete response was defined as the disappearance of all measurable lesions for at least 1 month. A partial response was defined as greater than 50% disappearance of all measurable lesions for at least 1 month. Stabilization was defined as less improvement of lesions than a partial response but no evidence of new lesions. Progressive disease was defined as the appearance of new skin lesions.
Adjunctive therapy Adjunctive therapy was considered on an individual basis depending on the severity of the disease and initial response to ECP. Adjunctive therapies were not used for the initial 4 to 6 months of the ECP treatment period. Adjunctive therapies included topical nitrogen mustard (10 mg/dl concentration in ointment), oral chlorambucil and prednisone, methotrexate, electron beam radiotherapy, PUVA, combination chemotherapy, and topical/intralesional steroids. The response to the agents used adjunctively was compared before and after ECP to judge additive or synergistic effects.
Statistical analysis To determine whether the clinical, morphologic, immunologic, or treatment variables would predict the final outcome, these were tested against the final outcome defined as either responders (complete or partial remission) or nonresponders (stabilization or progressive disease [see later]). Continuous variables were tested with a t test. Discrete variables were assessed with a chi-square test, or, for two-by-two tables, with Fisher's exact test. Criterion for significance was p < 0.05. Data are presented as mean + standard error of the mean. RESULTS
Patient population Table I summarizes the clinical data for the C T C L patients. The mean age at diagnosis was 61.2 __+ 3.2 years. There were 10 men and 10 women. The mean interval from initial symptom to
Photopheresis in CTCL 733 Table Ill. Morphologic and inamunologic studies Characteristic
No.
Extracutaneous disease Lymph nodes Bone marrow Peripheral blood Transformed CTCL T-cell immunophenotype Helper T-cell subtype Aberrant T-cell phenotype Expression of activation markers
8/20 4/6 5/19 6/19 5/20 18/ 18 16/18 7/18 15/16
diagnosis of CTCL was 5.9 _ 2.0 years. The mean interval from diagnosis of C T C L to onset of E C P was 2.2 +__0.4 years. Three patients (Nos. 8, 19, 20) showed improvement of their clinical staging with therapies before ECP, but were not free of disease. Patients 8, t9, and 20 were described as having clinical stages IVB, III, and IIA, respectively, at the time of diagnosis of CTCL. All other patients were considered to be at their worst clinical stage at the onset of ECP. At the start of ECP 14 patients were in the generalized patch/plaque stage (T2), two were in the tumor stage (T3), and three patients had exfoliative erythroderma (T4). Nix months before ECP, patient 18 had generalized patch/plaquestage disease and biopsy-confirmed bone marrow involvement with CTCL. After 6 months of topical nitrogen mustard therapy the patient's skin lesions improved dramatically (To). ECP was initiated at this stage because of persistent bone marrow involvement at the request of the referring oncologist because no other therapy had been effective and because of the previous success with his patient (8) who had marrow involvement.
Morphologic and immunologic studies The morphologic and immunologic characteristics are summarized in Table III. Eight patients had proven extracutaneous disease with involvement of the lymph nodes (six cases), bone marrow (five cases), and circulating Srzary cells more than 15% of lymphocytes (six cases). In five cases there was evidence of transformation from C T C L to the largecell lymphoma variant. 22 In t8 patients who had complete phenotyping performed, there were 16 T-cell helper subtypes and 7 aberrant T-cell subtypes. In 15 of 16 patients there was expression of T-cell activation markers. 26
Journal of the American Academy of Dermatology
734 Zic et aL Clinical response Eleven patients showed a partial or complete response to ECP. The disease stabilized in four patients and progressed in five patients. Ten of 14 patients with generalized patch/plaque stage responded. Neither of the two patients with tumorstage disease improved. One of the three patients with refractory exfoliative erythroderma responded. Patients were categorized as responders (partial or complete response) and nonresponders (progressive disease or stabiliTation) for the purpose of analysis. The mean age of the responding group was 59.3 + 3.2 years compared with 63.8 +__3.3 years for the nonresponding group (nonsignificant difference). The CTCL patients who responded did so usually within the first 6 months of therapy. Six patients (Nos. 4, 5, 6, 8, 9, and 13) were successfully weaned from ECP. One patient relapsed (No. 13) requiring reinitiation of ECP treatments 4 months after weaning. At this time, patients 4, 5, 6, 8, and 9 remain in clinical remission off ECP treatments for 4, 9, 9, 9, and 3 months, respectively. Patient 4 continues to receive methotrexate therapy. Patient 11 was weaned from ECP after only modest progressive improvement after 16 sessions. However, this patient had a marked progression of disease since weaning from ECP. Comparison of clinical data between responders and nonresponders showed no statistically significant differences between the two groups with regard to race, age, the interval between initial symptom and diagnosis of CTCL, the interval from diagnosis of CTCL to initiation of ECP, the stage of disease, the presence of extracutaneous involvement, previous therapies including electron beam therapy and chemotherapy, the average leukocyte fraction exposed to UVA light, or the mean psoralen serum levels achieved. Mean serum psoralen levels were 153 + 20 mg/ml (range 49 to 285 mg/ml). When analyzed by simple t test, the only variable significantly associated with outcome (response vs no response) was the number of ECP sessions (t9 = 0.025). To determine whether this difference was associated with other variables, logistic regression was also performed. With this approach, the number of ECP sessions was again the variable that best predicted outcome (.p < 0.05). In contrast, no additional or separate beneficial effect of adjunctive
chemotherapy (p > 0.5) or preceding electron beam therapy (p > 0.1) was found. Adverse effects of ECP More than 500 ECP treatments were performed with minimal patient complications or adverse effects. The most common side effect was nausea after ingestion of 8-MOP. The nausea was transient and rarely interfered with treatment. Other less common comphcations included hypotension responsive to fluid administration (three patients), hypoglycemia (two patients), and hematomas at catheter sites (three patients). One patient (No. 18) experienced hemolysis during an ECP session resulting in hemoglobinuria, but without a decline in renal function. An extensive evaluation revealed no identifiable cause for the episode other than mechanically induced hemolysis confirmed by Therakos after examination of the photocassette. This patient tolerated subsequent ECP treatments without complications. Adjunctive therapy ECP improved or stabilized the clinical status of 15 of 20 patients if judged by previous therapy status. In combination with ECP 10 of 20 patients were able to be maintained with minimal additional therapy: topical steroids alone (patients 5, 15, and 18); PUVA alone (patient 16); nitrogen mustard alone (patients 12, 13, and 20); topical steroids and nitrogen mustard (patient 7); topical steroids and methotrexate (patient 4); and topical steroids combined with chlorambucil and prednisone (patient 19). DISCUSSION The beneficial effect of ECP for CTCL patients with S6zary syndrome has been clearly established.20, 21, 26-29FoUow-up studies by Heald et al. 26 found that erythrodermic patients did better when ECP was started earlier in the course of the disease and before patients were heavily pretreated. Only one of the three heavily treated patients with erythrodermic CTCL in our series responded to ECP and required adjunctive therapy. The beneficial effect of ECP for established tumor-stage CTCL patients has been less encouraging. However, some patients reportedly have had favorable responses when EQP was combined with electron beam radiotherapy.21, 26 Both patients in
Volume 27 Number 5, Part 1 November 1992
this series with advanced tumor-stage disease who already had other conventional and experimental treatments showed no response to ECP even with adjunctive combination chemotherapy, PUVA, and external beam radiotherapy. W e report the largest series thus far of CTCL patients with generalized patch/plaque stage (T2) disease treated with ECP. Of the 14 patients in this series with generalized patch/plaque-stage disease, 10 had a favorable response to ECP. Four patients demonstrated a complete response and six had a partial response. Four of the 10 responding patients have been successfully weaned from ECP and remain in remission at this time. Most patients in this series received extensive therapy before ECP and all had either shown no response or had progression of disease during these therapies. It was unclear from the data whether certain previous therapies made CTCL patients more sensitive to the beneficial effects of ECP. Establishing the degree to which adjunctive therapy will affect the outcome of the patients in this series is also difficult. However, 10 of 20 patients were able to be maintained with only minimal adjunctive therapy, whereas they had unstable disease before ECP. Patients received different adjunctive therapies for a variable time after ECP in this series. Adjunctive therapy with ECP has not been systematically evaluated in the treatment of CTCL. Heald et al. 26 suggested that low-dose methotrexate may be the best adjunctive therapy with ECP because it is relatively nontoxic and nonimmunosuppressive. The patients in this series receiving methotrexate did not demonstrate any striking difference in response compared with those patients not receiving the drug. In our series, the only consistent therapeutic intervention during the period of observation in this series was ECP. Therefore we attributed a major portion of the patients' responses to this treatment modality but believe that adjunctive therapy is helpful. Multivariate analysis revealed that the only factor that correlated significantly with the final outcome was the number of ECP sessions the patients received. Because we did not systematically evaluate the effect of adjunctive therapy, we cannot precisely define the contribution of any single agent other than ECP to the final outcome. W e observed that four of five CTCL patients with stage IB disease (generalized patch/plaque, no elin-
Photopheresis in CTCL 735 icaUy abnormal lymph nodes) responded to ECP. This suggests that ECP may be important in the treatment of this subgroup of patients with disease apparently localized to the skin. We also observed that a patient with Ki-l-positive lymphoma is free of disease after failing to respond to radiation and bone marrow transplantation. Our experience with ECP supports the conclusion of others 19-21,26"29 that ECP is a relatively safe modality with infrequent complications. Complications associated with other treatment modalities for CTCL such as myelosuppression, fiver toxicity, renal toxicity, neural toxicity, and radiation dermatitis have not been reported with ECP. The hemolytic episode that occurred during an ECP session was probably due to mechanical hemolysis and not from ECP itself. The mechanism of the response to ECP has not been completely elucidated. Edelson noted that it appears to involve an induced immunologic reaction against the patient's malignant T cells. 19,28 Photoactivated 8-MOP forms covalent pfiotoadducts with nuclear DNA, thereby affecting the antigenicity and immunogenicity of exposed nucleated cells.19 Evidence from animal and human studies suggests that the reinfusion of the irradiated T cells induces a therapeutically significant immunologic reaction that targets the malignant clones. 19,28,29 Our studies did not address these mechanisms. REFERENCES
1. Edelson RL. CTCL: the Srzary syndrome, mycosis fungoides, and related disorders. Ann Intern Med 1975;83:53452. 2. Weinstock MA, Horrn JW. Mycosis fungoides in the United States: increasing incidence and descriptive epidemiology. JAMA 1988;260:42-6. 3. Weinstock MA, Horm JW. Population-based estimate of survival and determinants of prognosis in patients with mycosis fungoides. Cancer 1988;62:1658-61. 4. Rosenbaum MM, Roenigk H H Jr, Caro WA, et al. Photochemotherapy in cutaneous T cell lymphoma and parapsoriasis en plaques. J AM ACAD DERMATOL1985;13:61322. 5. Hrnigsmann H, Brenner W, Rauschmeier W, et al. Photochemotherapy for cutaneous T cell lymphoma. J AM ACAD DERMATOL 1984; 10:238-45. 6. RamsayDL, HalperinPS, Zeleniuch-JacquotteA. Topical mechlorethamine therapy for early stage mycosis fungoides. J AM ACAD DERMATOL 1988;19:684-91. 7. Vonderheid EC, Tan ET, Kantor AF, et al. Long-term efficacy, curative potential, and carcinogenicity of topical mechlorethamine chemotherapy in cutaneous T cell lymphoma. J AM ACAD DERMATOL 1989;20:416-28. 8. Cotter GW, Baglan R J, Wasserman TH, et al. PaUiative
736
9. 10. 11. 12. 13.
14. 15. 16. 17.
18.
Z i c et al.
radiation treatment of cutaneous mycosis fungoides: a dose response. Int J Radiat Oncol Bid Phys 1983;9:1477-80. Hoppe RT, Cox TS, Fuks Z, et al. Electron-beam therapy for mycosis fungoides: the Stanford University experience. Cancer Treat Rep 1979;63:691-700. Kuzel TM, Springer E, Roenigk HH Jr, et al. Mycosis fungoides: time to define the best therapy. J Natl Cancer Inst 1990;82:183-5. Grozea PN, Jones SE, MeKelvey EM, et al. Combination chemotherapy for mycosis fungoides: a Southwest Oneology Group study. Cancer Treat Rep 1979;61:279-82. Minna JD, Roenigk HH Jr, Glastein E. Report of the Committee of Therapy for Mycosis Fungoidez and S~zary Syndrome. Cancer Treat Rep 1979;63:729-36. Bunn PA Jr, Fischmann AB, Schechter GP, et al. Combined modality therapy with electron-beam irradiation and systemic chemotherapy for cutaneous T cell lymphoma. Cancer Treat Rep 1979;63:713-7. Braverman IM, Yager NB, Chen M, eta[. Combined total body electron beam irradiation and chemotherapy for mycosis fungoides. J AM ACAD DERMATOL1987;16:45-60. Knobler RM, Edelson RL. Cutaneous T cell lymphoma. Med Clin North Am 1986;70:109-38. OIsen EA, Rosen ST, VoUmer RT, et al. Interferon alfa-2a in the treatment of cutaneous T cell lymphoma. J AM ACAD DERMATOL 1989;20:395-407. Kaplan EH, Rosen ST, Norris DB, et al. Phase I[ study of recombinant human interferon gamma for treatment of cutaneous T cell lymphoma. J Natl Cancer Inst 1990; 82:208-12. Kuzel TM, Gilyon K, Springer E, et al. Interferon alfa-2a combined with phototherapy in the treatment of cutaneous T ceil lymphoma. J Natl Cancer Inst 1990;82:203-7.
Journal of the American Academy of Dermatology
19. Edelson RL. Photopheresis: a new therapeutic concept. Yale J Biol Med 1989;62:565-77. 20. Edelson R, Berger C, Gasparro F, et al. Treatment of cutaneous T cell lymphoma by extracorpereal photochemotherapy. N Engl J Med 1987;316:297-303. 21. Armus S, Keyes B, Cahill C, et al. Photopheresis for the treatment of cutaneous T cell lymphoma. J AM ACAD DERMATOL1990;23:898-902. 22. Salhany KE, Cousar JB, Greer JP, et al. Transformation of cutaneous T cell lymphoma to large ceU lymphoma: a clinicopathologic and immunologic study. Am J Pathol 1988; 132:265-77. 23. Borowitz M J, Croker BP, Burchette J. Immunocytochernicaldetection of lymphocytesurface antigens in fixed tissue sections. J Histochem Cytochem 1982;30:171-4. 24. Pinkus GS, Thomas P, Said JW. Leu-Ml--a marker for Reed-Sternberg cells in Hodgkin's disease: an immunoperoxidase study of paraffin-embedded tissues. Am J Pathol 1985;I 19:244-52. 25. Bunn PA, Lamberg SI. Report of the committee on staging and classification of cutaneous T cell lymphomas. Cancer Treat Rep 1979;63:725-7. 26. Heald PW, Perez MI, Christensen I, et al. Photopheresis therapy of cutaneous T cell lymphoma: the Yale-New Haven Hospital experience. Yale J Biol Med 1989;62:629-38. 27. Knobler RM. Photopheresis--extracorporeal irradiation of 8-MOP containing blood: a new therapeutic modality. Blut 1987;54:247-50. 28. Heald PW, Edelson RL. Photopheresis therapy of cutaneous T cell lymphoma. Keio J Med 1988;37:155-67. 29. Edelson RL. Photophercsis. J Clin Apheresis 1990;5:77-9.
