Phase II trial of interferon-α-2a plus psolaren with ultraviolet light A in patients with cutaneous T-cell lymphoma




To evaluate the efficacy and side effects of psolaren with ultraviolet light A (PUVA) and interferon-alpha-2a (IFN-α-2a) in patients with mycosis fungoides (MF) and Sézary syndrome (SS).


From May 1993 to January 1999, 63 symptomatic patients with all stages of MF and SS were treated in a prospective Phase II trial with systemic escalating doses of IFN-α-2a combined with PUVA for 1 year, followed by indefinite PUVA maintenance in complete responding patients.


Sixty-three patients were enrolled (Stage IA, n = 6; IB, n = 37; IIA, n = 3; IIB, n = 3; III, n = 12; IVA, n = 2). Ten patients had received previous therapy. The median follow-up duration for the entire cohort is 37 months. Of 63 patients, 51 achieved a complete response (CR; 74.6%) or partial response (PR; 6%) to therapy. The median response duration is 32 months. The 5-year overall survival rate is 91% and the 5-year disease-free survival rate is 75%. No life-threatening side effects were observed. Five patients stopped IFN-α-2a therapy due to toxicity. Eighty-four percent of the patients received more than 75% of the planned dose (12 million units three times a week).


This combination of IFN-α-2a and phototherapy is an effective and safe therapy for patients with symptomatic MF. Cancer 2002;95:569–75. © 2002 American Cancer Society.

DOI 10.1002/cncr.10706

Cutaneous T-cell lymphomas (CTCL) represent a heterogeneous group of non-Hodgkin lymphomas characterized by the malignant proliferation of helper T lymphocytes that infiltrate the skin.1 When first diagnosed, the disease is confined to the skin. Subsequently, it disseminates into the lymph nodes, blood, or viscera with the most commonly identified visceral sites being the lungs, spleen, and liver. Mycosis fungoides (MF) and Sézary syndrome (SS) are the most common forms of CTCL, comprising 80–85% and 10–15%, respectively, of this disease. The natural history of MF is characterized by an indolent progression through four stages: patch, plaque, tumor, and visceral involvement. The rate of disease progression is variable and it is possible for patients to have a simultaneous expression of patches, plaques, and tumors in different areas of their skin.2–4 A subset of patients may present with or evolve into SS, characterized by involvement of the blood by atypical T lymphocytes with hyperconvoluted cerebriform nuclei, termed Sézary cells, and an associated generalized exfoliative erythroderma.5

The number of atypical cells, or Sézary cells, necessary to define the syndrome, as well as the best method of identification of Sézary cells in the peripheral blood, remains controversial. Some researchers have used a cutoff of 5%, 9%, and 10% of the total leukocyte count as an indicator of the disease.6 The TNMB classification for MF and SS considers the type and extent of skin (T [tumor] classification), lymph node (N classification), visceral (M [metastasis] classification), and blood (B classification) involvement.2

The median survival time from diagnosis was 10–12 years for early-stage MF confined to the skin and less than 2 years7 for late-stage disease involving extracutaneous sites. Treatments, tailored to the stage of disease, are indicated to modulate symptoms, improve clinical appearance, and prevent progression of disease, which has an impact on survival.3

Therapy for CTCL has encompassed a variety of modalities, including topical and systemic chemotherapies,8, 9 topical corticosteroids,10 oral psolaren plus ultraviolet light (PUVA),11, 12 total skin electron beam therapy (TSEBT),13, 14 extracorporeal photochemotherapy (ECP; photophoresis),15 retinoids,16, 17 adenosine analogs,18, 19 monoclonal antibody (MoAb) therapy,20 and interferons (IFN).21 The use of IFN-α to treat CTCL patients was first reported by Bunn et al. in 198422 to be effective in the treatment of advanced and heavily pretreated patients. Usual doses, mostly given subcutaneously, range from about 3 million units (MU) three times weekly (TIW) to 9 or 12 MU daily.23 In a metaanalysis of 304 patients, the overall response rate was 70% with the best responses in the early stages.22 In preliminary studies, IFN was used at high doses for short periods showing a dose-related toxicity. In spite of this, a clear correlation between dose and efficacy has never been confirmed.24 The combination of IFN-α plus PUVA showed a higher response rate in Phase II studies and a longer response duration than PUVA or IFN-α alone25–28 and was superior to IFN-α plus acitrein in a randomized study.29 The optimal IFN-α dose, schedule, and duration, whether used alone or combined with PUVA, is still a critical issue. Moreover, the efficacy of this combination in patients with the advanced stage of the disease is still not clear. To assess the efficacy and tolerability of low IFN-α doses plus PUVA in the treatment of MF and SS patients, we started this prospective clinical trial conducted in the Departments of Dermatology and Oncology, respectively, of Padua and Rovigo Hospitals.



Patients of at least 18 years of age of either gender with MM or SS, Stage IA–IVA, seen in the Department of Dermatology and Oncology, respectively, of Padua and Rovigo Hospitals, from May 1993 to January 1999, were enrolled in this study. Patients with Stage IA were enrolled only if they had more than one lesion. The protocol was revised and approved by the Scientific Committee of the Veneto Lymphoma Group. Informed consent was obtained from each patient before enrollment, in accordance with the rules and regulations of the Ethics Committee of Padua Oncology Department Eligibility requirements included all the following: 1) histopathologic evidence of MF or SS, 2) Eastern Cooperative Oncology Group performance status of 2 or less, 3) preserved hepatic, renal, and hematologic function, 4) age older than 18 years, 5) absence of active infection, 6) no previous systemic chemotherapy or TSEBT, 7) no history of previous neoplasm, 8) absence of psychiatric disorders, and 9) absence of serious heart disease, seizure disorders, or dysplastic nevi.

All histopathology was performed independently by two pathologists from the Padua Pathology Department (A.P. and F.M.). The diagnosis of MF was based on light microscopic evidence of a dermal infiltrate of atypical lymphocytes with cerebriform nuclei, epidermal exocytosis of these cells, and the presence of microabscesses. SS diagnosis was based on detection by light microscopic examination of blood smears, of more than 5% of circulating atypical lymphocytes with hyperconvoluted nuclei. Immunohistochemistry was added on frozen sections with the following antibodies: CD2, CD3, CD4, CD7, and CD8. In selected cases, diagnoses were confirmed by Southern blot analysis of the β chain gene of the T-cell receptor to determine whether a clonal rearrangement was present.

Staging procedure included history and physical examination, a computed tomography scan of the chest, abdomen, and pelvis, bone marrow aspiration and biopsy, complete blood count with differential and platelet count, a serum screening chemistry panel, antinuclear autoantibodies, and ocular examination. Lymph node biopsies were performed in patients who had palpable lymphoadenopathy before treatment. The stage was determined according to the criteria proposed by the TNMB staging system for CTCL.1, 2 Physical examination and blood counts were repeated every month for the first 3 months, every 2 months in the first year of the treatment, every 3 months in the second year, and every 6 months thereafter. Patients were evaluated both by a dermatologist and a hematooncologist.


Treatment was initiated with increasing doses of IFN-α-2a (Roferon-Hoffmann-La Roche, Basel, Switzerland). Patients were subcutaneously administered 3 MU TIW in the first week, 6 MU three times per week (TIW) in the second week, 9 MU t.t.w in the third week, and 12 MU TIW from the fourth week to the first year. The escalation was done only if no side effects (Grade ≥ 2) were observed during the treatment. Photochemotherapy was performed with oral 8-methoxipsolaren 0.5 mg/kg 2 hours before UVA radiation, three times per week, until clinical complete skin clearing, once a week for 4 weeks, and then indefinitely once every 2–4 weeks. UVA doses started with 1.5 J/cm2 and were increased by 0.5 J/cm2 up to the minimal erythema dose.

Toxicity was scored according to the Common Toxicity Criteria established by the National Cancer Institute. If a Grade II hepatic, neurologic, hematologic, or constitutional toxicity was observed, the INF dose was reduced by one level (25%). If a Grade III or higher cardiac, cutaneous, and endocrinologic toxicity was observed, the treatment was stopped for 2 weeks and then resumed, in the case of recovery, with a 25% reduction.

CR was defined as disappearance for at least 4 weeks, of all skin and lymph node disease by physical examination, in selected cases confirmed by skin biopsy, with absence of circulating atypical lymphocytes on peripheral smear. PR was defined as a minimum of 50% reduction of skin or lymph node disease as determined by the sum of the products of the largest perpendicular diameters of all measurable lesions, with no evidence of new lesions, for at least 4 weeks. No response (NR) was defined as any result less than a PR. Progressive disease (PD) was defined as a more than 25% increase of skin or lymph node disease or as the appearance of new lesions.

To graduate the clinical response in SS patients, we used the reduction of the skin score, tabulated as the sum of the product of severity (0 = normal, 1 = barely detectable erythema and scaling, 3 = marked erythema and exfoliation, and 4 = fessuring, maximal erythema, induration, and tumors) and surface area percentage.30 To define a CR, complete skin clearance for at least 4 weeks was required as was the absence of circulating Sézary cells on peripheral smear.

Relapse-free survival (RFS) was defined as the time from the documentation of CR to the presentation of recurrence and overall survival (OS) was defined as the time from the start of treatment to the last control. RFS and OS were analyzed using the Kaplan–Meier product-limit curve. All data were completed and analyzed in September 2000.


The characteristics of the 63 patients enrolled in this study are listed in Table 1. There were 42 males and 21 females. The median age was 62 years (range, 23–85) with 19 patients (30%) being 70 years or older. Of the 63 patients, 61 had a diagnosis of MF and 2 had a diagnosis of SS. Of them, 37 (59%) were Stage IB and 10 (16%) had been pretreated. The pretreatment consisted of UVA alone in two patients, PUVA in two, limited-field radiotherapy in one, IFN alone in one, and steroids in four patients.

Table 1. Patient's Characteristics
CharacteristicsNo. of patients (%)
No. of patients63 (100)
 Male42 (67)
 Female21 (33)
Age (yrs) 
 >70 yrs old19 (30)
Prior treatment 
 No53 (84)
 Yes10 (16)
Stage of disease 
 IA (T1N0)6 (9)
 IB (T2N0)37 (59)
 IIA (T1–2N1)3 (4)
 IIB (T3N0–1)3 (4)
 III (T4N0–1)12 (19)
 IVA (T1–4N2–3)2 (3)

Of the 63 patients, 47 (74.6%; 95% confidence interval: 56.3–92.9%) obtained a CR, 6 obtained a PR, 2 were classified as NR, and 5 had PD. The median time to obtain the complete remission was 7 months (range, 1–17); the median duration of overall response was 32 months, with a range of 6–57 months. Three patients stopped the treatment in the first month: one due to neurologic intolerance to IFN and one due to a generalized symptomatic papular skin eruption in the first week of therapy. This skin reaction improved after the withdrawal of IFN and recurred when it was resumed. The third patient stopped the treatment due to a gastrointestinal intolerance to 8-methoxypsolaren (Table 2). CRs were obtained in all stages of disease. Five PDs were observed in both patients with SS (Stage III and IVA) and in another three patients (two in Stage IB and one in Stage III). Nine patients had a recurrence at a median time of 12 months (range, 6–27): five of them obtained a second CR with IFN and PUVA therapy, three obtained a CR with UVA alone, and an 80-year-old patient refused further treatment. Forty-seven (74.6%) patients are alive without disease at a median follow-up time of 37 months (range: 4–88) and 12 (19%) are alive with disease. The 5-year OS for the entire cohort is 91.3% (Fig. 1) and the 5-year relapse-free survival (RFS) rate is 75.1% (Fig. 2). The median DFS is 32 months, and it is significantly different between Stages IA-IB-IIA vs. Stages IIB-III-IV (P < 0.05) (Fig. 3). Four patients died, one of whom had progressive MF (Stage T2N0M0). Two nonresponding patients died, one with Stage T4N1M0 MF for a lung carcinoma after 26 months and the other of myocardial failure after 22 months. This patient had stopped IFN treatment 9 months before for Grade III flulike symptoms. The fourth patient in CR from MF died of an anaplastic astrocytoma after 12 months. Eighty-four percent of patients received 75% or more of the planned IFN dose. Fourteen patients (22%) tolerated the full planned dose for the whole treatment period. Only four patients (6%) received less than 50% of the planned dose (Table 3).

Table 2. Results
ResponseNo. of patients (%)
  1. CR: complete response; PR: partial response; NR: no response; PD: progressive disease; MF: mycosis fungoides; SS: Sézary syndrome.

CR47/63 (74.6)
PR4/63 (9)
NR2/63 (3)
PD5/63 (8)
Not evaluable3/63 (4)
Recurrences9/47 (19)
Alive without disease47 (74.6)
Death4 (6)
Alive with MF/SS12 (19)
Figure 1.

Overall survival (months from start of the chemotherapy).

Figure 2.

Disease-free survival (months from start of the treatment).

Figure 3.

Relapse-free survival according to stage (months from start of the therapy).

Table 3. Percentage of Interferon Administered
No. of patients (%)14 (22)38 (60)7 (11)4 (6)

Life-threatening adverse events were not observed. The most common adverse symptoms (Table 4), requiring dose reduction were fatigue, neuromood and neuromotor disorders, elevation in liver enzymes, dryness with desquamation of the skin, weight loss, thrombocytopenia, and leukopenia. In some patients, hypothyroidism due to antithyroglobulin auto antibodies, impotence, hair loss, cardiac arrhythmias and elevation in triglycerides were also observed. IFN was withdrawn due to toxicity in five patients.

Table 4. Toxicity
SymptomsNCI-CTC grade/patient (N = 63)a
  • NCI-CTC: National Cancer Institute - Common Toxicity Criteria.

  • a

    Data are expressed as the number of patients. Some patients had more than one adverse reaction.

Flulike symptoms43
Skin dryness/redness51
Weight loss21
Cardiac dysarrhythmia01
Hair loss11


A number of treatment options such as topical nitrogen mustard, topical carmustine, PUVA, TSEBT, ECP IFN, adenosine analogs, retinoids, and MoAbs are available for the treatment of patients with CTCL.8 In spite of the many therapies available, there is limited evidence to show that MF can be cured.26 Topical chemotherapy,9 PUVA,11, 12 and TSEBT13 have established roles in the palliation of the early stages of disease, but advanced disease is less effectively treated with these modalities. Moreover, neoplastic clones are detectable at distant sites and in the blood even in the presence of clinically localized skin disease.7, 33 Combined radiation plus chemotherapy results in a greater CR rate than sequential topical single agents alone. However, no survival advantage has been documented with the early use of aggressive regimens.31, 32

CTCLs have an indolent course. Nevertheless, the quality of life is impaired by symptoms associated with skin involvement. Therapeutic goals should focus on symptom relief through durable remissions without excessive acute and late toxicity.

Recombinant IFN-α is the cytokine with which the greatest experience has been obtained. Clinical response rates have been reported for 50–80% of treated patients receiving systemic administration via subcutaneous or intramuscular routes. Responses have even been obtained in patients with advanced and heavily pretreated disease.27

Some very promising results have been reported using the combination of IFN-α and PUVA. Kuzel et al.25 reported a 80% CR rate with a median duration that approached 2 years for 15 patients treated with a combination of IFN-α2-a (6–30 MU TIW) and PUVA. Mostow et al.26 reported five CRs in five patients refractory to PUVA alone, with the addition of IFN-α2-a 3–6 MU daily. In a subsequent study including 39 patients, 87% pretreated and 36% with advanced disease, Kuzel et al.27 reported a 62% CR rate with a median duration of 28 months.

In the 63 patients treated, we obtained a 75% CR rate with a median duration of 32 months. Furthermore, of the nine patients with recurrence (19%), five obtained a second CR with the same therapeutic scheme. Overall, 68% of treated patients are alive without disease at a median follow-up of 34 months. These results confirm that the CR rates obtained with the combination of IFN and PUVA are better than those reported for IFN-α or PUVA alone. We also confirm that the CR is achievable in all stages of the disease. Conversely, we did not observe any responses in SS patients. However, in our series, only two patients had SS. The differences observed in overall survival by Kuzel et al.27 between early-stage and advanced-stage patients were not supported by our data. In our study, 3 of 14 patients in advanced stage of disease developed a second neoplasia and 7 of the 11 remaining are alive without disease at a median follow-up of 40 months.

An apparent increase in second malignancies has been recognized in other reports34 and were considered to be CTLC related, rather than treatment related. However, in our series, they were believed to be age related.

In the study reported by Kuzel et al.,27 48% of patients needed a reduction in IFN dose due to toxicity and constitutional symptoms were experienced by 85% of patients. Most of the side effects of IFN-α (fever, myalgias, chills, fatigue, malaise, anorexia, weight loss, metallic taste) were dose related, although the dose of IFN-α that causes side effects has great interindividual variability.24 Initiation with lower doses and escalation to the planned level over weeks reduce the flulike symptoms.27 In our study, we chose a dose escalation scheme of IFN because there is not clear correlation between dose and response, but there is between dose and toxicity. Using a lower dose than Kuzel et al.27 and an escalating scheme, we maintained a high complete remission rate and limited toxicity. Only two patients stopped the treatment due to IFN-α toxicity, whereas 82% of patients received 75% or more of the planned dose, even if 30% of them were older than 70 years old. Stadler et al.29 obtained the same results in a randomized study that compared 9 MU TIW of IFN-α plus PUVA versus IFN-α plus acitretin. Only 10% of patients experienced severe side effects (World Health Organization Grade 3) and 2 of 40 had to be withdrawn from the study due to toxicity. In this study,28 the combination of IFN-α plus PUVA obtained a significantly better complete remission rate than the combination of IFN-α plus acitretin.

It is not clear from the literature and from our data whether a prolonged administration of IFN-α could prolong the duration of complete remission or not. Given the side effects and the cost of therapy, and considering that in our study the median time to obtain a CR was 7 months, we believe that it is not essential to prolong IFN therapy for more than 1 year. Conversely, as in the study by Kuzel et al.,27 we believe it is useful to maintain PUVA therapy indefinitely.

A randomized study is needed to establish whether the combination of IFN-α plus PUVA is statistically better than PUVA alone and whether this schedule should be applied to all stages of disease. Besides considering the long survival of patients with CTCL and the disturbing symptoms related to the active disease, the endpoint of this study must evaluate not only the failure-free and OS but also the quality of life and the median time to achieve the CR. We believe that IFN-α combined with PUVA is an effective and safe therapy for patients with symptomatic MF.


The authors thank Ms. Denise Kilmartin for her help with this study.