Interferon-α (IFN-α) and retinoids have shown nonoverlapping toxicity and each has shown antitumor activity in patients with lymphoma. The aim of the current study was to assess the toxicity, safety, and efficacy of IFN-α combined with isotretinoin in patients with advanced, refractory lymphoid malignancies.
Adults with biopsy-proven advanced lymphoid malignancy were treated. Patients with compromised bone marrow function (platelet counts as low as 30 × 109/L) were eligible. Treatment was comprised of IFN-α at a starting daily dose of 3 mega units subcutaneously and isotretinoin orally starting at a dose of 1 mg/kg daily in 2 divided doses.
Forty-four patients were evaluable. Their median age was 57 years (range, 18–82 years). Eighteen patients had advanced cutaneous T-cell lymphoma, 6 patients had peripheral T-cell lymphoma, 14 patients had Hodgkin disease, and 6 patients had a variety of other lymphoid malignancies. Patients with Hodgkin disease had received a median of 6 previous therapies (range, 3–12 therapies) and patients with other lymphoid malignancies had received a median of 4 previous therapies (range, 1–9 therapies). The median duration of treatment was 4 months (range, 0.25–38 months). The overall response rate was 38.6% (complete response in 5 patients [11.3%] and partial response in 12 patients [27.3%]). The median response duration was 3 months (range, 1–95+ months). The most common toxicities were low-grade fever, flu-like symptoms, and fatigue (IFN-α effects); dry mouth and skin and hypertriglyceridemia (cis-retinoic acid effects); and thrombocytopenia (which generally occurred in patients with low baseline platelet counts).
In recent years, important advances have been made in the treatment of lymphomas. Several novel approaches, such as monoclonal antibody-based therapy and purine analogs, are promising. However, patients continue to experience disease recurrence even after intensive regimens of combination chemotherapy.1 The therapeutic options for recurrent lymphoma include regimens with low toxicity to more arduous therapies, such as stem cell transplantation (SCT). Despite the availiability of multiple therapies, a subset of patients eventually develop disease recurrence and die of their disease.
Of several bioimmunotherapeutic approaches, recombinant interferons (IFN) have shown antitumor activity against some malignant lymphomas, particularly the indolent B-cell lymphomas and the cutaneous T-cell lymphomas (CTCL).2 Responses to IFNs also have been observed in studies of patients with Hodgkin disease.3–6 Retinoids induce responses in approximately 50% of patients with CTCL.7–10 In addition, in a pilot study, 6 of 12 patients with peripheral T-cell lymphoma (PTCL) responded to 13-cis-retinoic acid (5 patients achieved a complete disease remission [CR]).11
In the current study, we administered IFN-alpha (IFN-α) in combination with isotretinoin to patients with refractory or recurrent non-Hodgkin lymphomas as well as to patients with Hodgkin disease who did not respond to two or more regimens of multiagent cytotoxic chemotherapy. The rationale for using IFN-α in conjunction with retinoids derives from the reported activity of this approach in studies of patients with CTCL,12–17 the principle of combining active agents with nonoverlapping toxicity profiles, and the finding that these drugs were minimally myelosuppressive and could therefore be administered to patients with compromised bone marrow reserve. The results of the current study demonstrate that the combination of IFN-α and isotretinoin has antitumor activity and is well tolerated in a group of patients with advanced, refractory lymphoid diseases.
MATERIALS AND METHODS
Eligibility criteria included histologic proof of incurable non-Hodgkin lymphoma or recurrent Hodgkin lymphoma confirmed at the M. D. Anderson Cancer Center. Patients with non-Hodgkin disease were treated only if they had failed conventional therapy and patients with Hodgkin disease were treated if they had failed at least two conventional multiagent chemotherapy regimens. Patients should not have received chemotherapy, immunotherapy, hormonal therapy, or radiotherapy within 3 weeks of entry into the study and must have recovered from the acute toxic effects of previous therapy. Age > 18 years, life expectancy of ≥ 3 months, and a performance status of ≤ 2 (Zubrod scale) were required. Patients signed an informed consent according to institutional procedures. Adequate hepatic function (bilirubin level ≤ 2.0 mg/dL and alanine aminotransferase level ≤ 4 times the upper limits of normal), adequate renal function (creatinine level ≤ 1.5 mg/dL), and serum triglyceride level ≤ 2.5 times the upper limits of normal were required. Patients with compromised bone marrow function, including platelet counts as low as 30 × 109/L, were eligible. Patients were eligible regardless of the extent of previous chemotherapy.
The pretreatment staging evaluation included a history and physical examination (including a neurologic examination and measurement of lymph nodes and skin lesions), a complete blood count with differential and platelet counts, coagulation tests, renal and hepatic function tests, triglyceride levels, bone marrow aspiration, biopsy, cytogenetics, immunophenotype, and molecular tests for T-cell receptor rearrangement. A chest radiograph and computerized tomography scans of the abdomen and pelvis were performed. All patients of child-bearing age were advised of the need for adequate contraception and of the teratogenicity of isotretinoin. Premenopausal women were screened for pregnancy using serum β-human chorionic gonadotropin.
Patient Monitoring during Therapy
During the first 12 weeks, patients were monitored at least once weekly with a complete blood cell count with differential. A biochemical profile serial multiple analysis-12 (SMA-12), triglyceride levels, and coagulation tests were performed at least every 2 weeks. Appropriate radiologic tests as well as bone marrow aspirations and biopsies in selected patients were repeated every 1–2 months. Bone surveys were performed every 1–2 years to determine changes.
Patients received 3 mega units (MU) daily of recombinant IFN-α (Intron® A; Schering Corporation, Kenilworth, NJ) subcutaneously and 1 mg/kg of isotretinoin (total daily dose) orally in 2 divided doses for 8 weeks. Patients without significant side effects who did not demonstrate a response after 4 weeks of therapy were escalated to receive 5 MU daily. Responders continued on therapy indefinitely or for at least 6 months after a CR had been achieved.
Patients with Grade 1 or 2 toxicities related to IFN-α were treated with the same dose or received the same dose three times per week instead of daily. If Grade 3 toxicity occurred, the dose of IFN-α or isotretinoin was reduced by 50%. If Grade 4 toxicity occurred, therapy was discontinued. The dose of IFN-α was reduced for toxicities consisting of fever, chills, fatigue, or somnolence. The dose of isotretinoin was reduced for toxicities consisting of mucocutaneous problems or hypertriglyceridemia. For other toxicities, both agents were reduced.
Duration of therapy
Patients were treated for 2 months, unless there was rapidly progressive disease, and for at ≥ 3 months, if there was no change in disease and if no Grade 3–4 toxicities occurred. Patients who achieved a CR continued to receive therapy for at least an additional 6 months and other responding patients until progression of disease occurred.
End Points and Statistical Methods
A CR was defined as the disappearance of all symptoms and signs of disease for ≥ 4 weeks. In patients with bone marrow involvement, an absolute neutrophil count of > 1 × 109/L and a platelet count of > 100 × 109/L were also required.
Partial response/remission (PR) indicated a ≥ 50% decrease in the sum of the products of the dimensions of all lesions for ≥ 4 weeks. For patients with bone marrow involvement, a 50% decrease in malignant lymphoid infiltrate was also required. Patients who did not meet the criteria for PR or progressive disease were denoted as having no response. Any increase > 25% in the sum of the products of the dimensions of any measurable lesion, the appearance of new lesions, growth of any lesion, or an increase in bone marrow lymphoid infiltrates indicated progressive disease. Response duration were measured from the time of response until there was evidence of progressive disease. For patients with CTCL, skin scores and response were assessed as previously published.18
Adverse events were graded as per the National Cancer Institute (NCI)-Common Toxicity Criteria.19 Progression-free survival was defined from documentation of response to disease recurrence.20 Progression-free survival curves were estimated by the Kaplan–Meier method.21 Patients were censored at the time of last follow-up.
Of 54 registered patients, 44 patients were evaluable. Ten patients were ineligible for the following reasons: errors in original diagnosis (n = 2 patients); medication error or noncompliance because patient did not receive isotretinoin (n =1 patient) or IFN-α (n =1 patient); registration error (n =1 patient); lost to follow-up (n =3 patients); or insurance issues necessitating early discontinuation of medication for financial reasons (n =2 patients).
The pretreatment characteristics of the patients are summarized in Table 1. Fourteen patients (32%) were age > 60 years. Of 37 patients, 23 (62%) had elevated levels of β-2-microglobulin (> 2 mg/L). Fifteen patients (34%) had elevated levels of lactate dehydrogenase (above the upper limit of normal).
Table 1. Patient Characteristics in 44 Patients with Lymphoid Malignancies Treated with Interferon-α and Isotretinoin
Hb: hemoglobin; PLT: platelets; LDH: lactate dehydrogenase; HTLV-1: human T-cell lymphotropic virus type 1.
Upper limit of normal of LDH at the study institution is G18 IU/L.
Patients with Hodgkin disease were especially heavily pretreated. They had received a median of 6 previous therapies (range, 3–12 therapies). For example, 79% had received previous therapy with doxurubicin, bleomycin, vinblastine, and dacarbazine or variants; 71% had received previous therapy with mechlorethamine, vincristine, procarbazine, and prednisone or variants; 57% had received previous therapy with etoposide, methylprednisolone, cytarabine, and cisplatin (ESHAP) or variants; and 36% had failed previous SCT.
Patients with non-Hodgkin lymphoma had received a median of four previous therapies (range, one to nine therapies). Sixteen patients (53%) had received previous combination cytotoxic chemotherapy regimens (Table 2). Thirty percent of the patients had received cyclophosphamide, doxorubicin, vincristine, and prednisone; 10% had received cyclophosphamide, vincristine, and prednisone; 10% had received ESHAP or variants; 3% had received sodium mercaptoethanesulfonate, ifosfamide, mitoxantrone, and etoposide; and 10% had received high-dose methotrexate, etoposide, and dexamethasone. Approximately 10% of the patients had received previous IFN-α alone, 20% had received a retinoid alone, and 23% had received previous total-body electron beam radiation. Individual patients had also received other chemotherapeutics, including but not limited to fludarabine and gemcitabine.
Table 2. Previous Therapy in 44 Patients with Lymphoid Malignancies Treated with Interferon-α and Isotretinoin
No. of patients
ABVD: doxurubicin, bleomycin, vinblastine, and dacarbazine; MOPP: mechlorethamine, vincristine, procarbazine, and prednisone; ESHAP: etoposide, methylprednisolone, cytarabine, and cisplatin; BEAM: carmustine, etoposide, cytarabine, and melphalan; CHOP: cyclophosphamide, doxorubicin, vincristine, and prednisone; COP: cyclophosphamide, vincristine, and prednisone; MINE: sodium 2-mercaptoethane sulfonate, ifosfamide, mitoxantrone, and etoposide; CMED: high-dose methotrexate, etoposide, and dexamethasone.
These patients had demonstrated an initial response to interferon alone, but then experienced disease progression.
The median duration of treatment was 4 months (range, 0.25–38 months). The median weekly dose of IFN-α was 21 MU. Seventeen patients required a dose reduction of IFN-α for the following reasons: thrombocytopenia (n = 6 patients); granulocytopenia (n = 2 patients); flu-like syndrome and atrial fibrillation (n = 1 patient); rash (n = 1 patient); pancreatitis, dizziness, and fatigue (n = 1 patient); extraordinary fatigue (n = 1 patient); muscle pain and depression (n = 1 patient); low tolerance (n = 3 patients); and unknown reasons (n = 1 patient).
The median daily dose of isotretinoin was 1 mg/kg. Seven patients received a reduction in the daily dose of isotretinoin for the following reasons: elevated levels of triglycerides (two patients), including one patient with concomitant cirrhosis; fatigue (one patient); poor compliance (two patients); patient's request (one patient); and unknown reason (one patient).
Response to Therapy
The overall response rate was 38.6%. Five patients achieved CR (11.3%) and 12 patients PR (27.3%). Table 3 shows the response by diagnosis. One patient previously treated with retinoid alone without adequate response responded to the combination therapy.
Table 3. Response by Diagnosis/Stage of Disease
CR + PR (%)
CR: complete response; PR: partial response; HTLV-1: human T-cell lymphocytic virus type 1.
Sezary syndrome/mycosis fungoides
Peripheral T-cell lymphoma
Diffuse large-cell lymphoma
HTLV-1 adult T-cell leukemia/lymphoma
Anaplastic large cell lymphoma
The median time to progression among responders was 5.7 months (range, 1–114+ months) (Fig. 1). The median time to progression in patients who achieved a CR was 6 months (range, 3–95+ months). In particular, the duration of response in these patients was 3 months, 4 months, 6 months, 33 months, and 95+ months. The median time to progression in patients who achieved a PR was 4.3 months (range, 1–114+ months).
Patients commonly manifested side effects typical of IFN-α (low-grade fever, flu-like symptoms, fatigue) or isotretinoin (dry mouth and skin; Table 4). Approximately 50% of the patients developed isotretinoin-induced hypertriglyceridemia that generally was controlled by the administration of gemfibrozil (at a dose of 600 mg orally twice daily). Grade 3–4 toxicities included the following: thrombocytopenia (n = 7 patients[16%], mostly patients with low baseline platelet counts); myocardial infarction (n = 1 patient); muscle pain and depression (n = 1 patient); nausea and weight loss (n = 1 patient); conjuctivitis and fatigue (n = 1 patient); renal failure (n = 1 patient); and skin reaction (n = 1 patient).
Table 4. Toxicity
No. of patients
No. of patients
NA: not applicable.
Ten of 12 patients who developed significant thrombocytopenia had low baseline platelet counts (< 100 × 109/L).
Myocardial infarction (Grade 4) in one patient; atrial fibrillation and pericarditis in one patient.
Injection-site reaction refers to skin rash and ulcerative skin lesion at the site of interferon-α injection.
We evaluated the combination of IFN-α and isotretinoin in patients with advanced refractory lymphomas. Retinoids are vitamin A analogs with anticarcinogenic activity. Their mechanism of action is complex and involves differentiation and apoptosis pathways.11, 22 In vitro data suggest synergism between IFN-α and retinoids; the combination induces higher levels of IFN-stimulated genes than either agent alone.23–26 In addition, retinoids augment the antiproliferative activity of IFNs23 and IFN-α may potentiate retinoid-induced differentiation.24, 26
The overall response rate to IFN-α and isotretinoin combination therapy was 39%. It is interesting to note that responses were observed in 4 of 6 patients (67%) with PTCL and 2 of these patients remained progression free 8 years and 9.5 years, respectively, after documentation of response. In addition, 6 of 18 patients (33%) with CTCL responded, as did 4 of 14 patients (29%) with refractory Hodgkin disease. These patients were heavily pretreated. The patients with PTCL had received a median of three previous therapies, the patients with Hodgkin disease had received six previous therapies, and the patients with CTCL had received four previous treatments. The role of the individual agents in inducing response could not, however, be inferred from the current study.
Among patients with CTCL in the current study, the response rate to the combination of IFN-α and isotretinoin (33%) was lower than that previously reported (approximately 50–60%) using IFN-α combined with a variety of retinoids.12–16 This may be because of the advanced stage of patients in the current study (Stage IIB-IV disease) and their heavily pretreated status. Many of these patients had received previous regimens containing multiagent cytotoxic chemotherapy. Regardless, these results confirm that this combination is active in patients with advanced and recurrent CTCL.
The current study data are consistent with results previously published in a few case reports that demonstrated the activity of IFN-α combined with a retinoid in non-CTCL malignancies. For instance, in a patient with Ki-1+ anaplastic large cell lymphoma (ALCL) refractory to cytotoxic chemotherapy, an IFN-α and 13-cis-retinoic acid combination regimen significantly reduced the size of a retroperitoneal mass.27 In another patient with Stage IV CD30-positive ALCL with brain, lung, and skin involvement, a combination of oral acitretin and IFN-α induced response in lesions at all involved sites.28
The median dose of IFN-α administered in the current study was 21 MU weekly. This dose was well tolerated by most patients. We included patients with poor bone marrow reserve and platelet counts as low as 30 × 109/L because this was believed to be a less myelosuppressive regimen than cytotoxics. Nevertheless, virtually all patients with pretreatment platelet counts ≤ 50 × 109/L had dose-limiting exacerbation of thrombocytopenia. Patients with baseline platelet counts > 50 × 109/L fared better. High-dose IFN-α (50 MU/m2 given subcutaneously 3 times per week) in patients with CTCL has been previously used in a Phase II study, but was considered to be above the maximally tolerated dose.29, 30 Similarly, a study by the NCI also found that high-dose IFN-α (10 MU/m2 on Day 1 followed by 50 MU/m2 on Days 2–5 every 3 weeks) was toxic.31 In two randomized studies comparing low-dose (3 MU intramuscularly daily) with high-dose IFN-α (dosage was escalated to 36 MU daily) in patients with CTCL32 or indolent lymphoma (3 MU subcutaneouly daily vs. 50 MU given subcutaneouly 3 times weekly),33 there was no advantage noted with regard to response rates or response duration for the higher doses.
IFN-α has been also investigated in Hodgkin disease.3–6 A study by the American Cancer Society demonstrated that IFN-α has no therapeutic benefit in these patients.6 However, some responses were observed in subsequent studies with small numbers of patients.3–5 In vitro studies of retinoids in Hodgkin disease have also been reported.34, 35 One study demonstrated the sensitivity of two Hodgkin lymphoma cell lines to isotretinoin.34 Proteomic analysis has shown that Reed-Sternberg cells possess molecular pathways that may be targeted by retinoids.35 In the current trial, 29% of patients with Hodgkin disease responded. These patients were heavily pretreated and were generally placed on trial because their exhausted bone marrow could no longer tolerate myelosuppressive chemotherapy.
The results of the current study confirm the activity of IFN-α and isotretinoin in patients with CTCL and demonstrate that this combination also has efficacy in patients with other lymphoid malignancies. The observation that patients with refractory Hodgkin disease and PTCL respond to this bioimmunotherapeutic approach is encouraging and warrants further investigation.