Multicenter phase 1/2 study of forodesine in patients with relapsed peripheral T cell lymphoma.

Peripheral T cell lymphomas are an aggressive group of non-Hodgkin lymphomas with poor outcomes for most subtypes and no accepted standard of care for relapsed patients. This study evaluated the efficacy and safety of forodesine, a novel purine nucleoside phosphorylase inhibitor, in patients with relapsed peripheral T cell lymphomas. Patients with histologically confirmed disease, progression after ≥ 1 prior treatment, and an objective response to last treatment received oral forodesine 300 mg twice-daily. The primary endpoint was objective response rate (ORR). Secondary endpoints included duration of response, progression-free survival (PFS), overall survival (OS), and safety. Forty-eight patients (median age, 69.5 years; median of 2 prior treatments) received forodesine. In phase 1 (n = 3 evaluable), no dose-limiting toxicity was observed during the first 28 days of forodesine treatment. In phase 2 (n = 41 evaluable), the ORR for the primary and final analyses was 22% (90% CI 12-35%) and 25% (90% CI 14-38%), respectively, including four complete responses (10%). Median PFS and OS were 1.9 and 15.6 months, respectively. The most common grade 3/4 adverse events were lymphopenia (96%), leukopenia (42%), and neutropenia (35%). Dose reduction and discontinuation due to adverse events were uncommon. Secondary B cell lymphoma developed in five patients, of whom four were positive for Epstein-Barr virus. In conclusion, forodesine has single-agent activity within the range of approved therapies in relapsed peripheral T cell lymphomas, with a manageable safety profile, and may represent a viable treatment option for this difficult-to-treat population.

Forodesine in the treatment of cutaneous T-cell lymphoma.

INTRODUCTION: Cutaneous T-cell lymphoma (CTCL) is characterized by the accumulation of neoplastic CD4+ T lymphocytes in the skin. Given the lack of curative treatments for CTCL, there is a significant need for new, superior therapies. Forodesine is a transition-state analogue that inhibits purine nucleoside phosphorylase. Because it selectively targets T lymphocytes, it represents a drug of interest for the treatment of CTCL. Areas covered: Phase I/II dose-ranging studies of intravenous (IV) and oral forodesine demonstrated its activity, safety, and tolerability for refractory CTCL. Response rates were 31% and 27%, respectively. No dose-limiting toxicities were observed. These studies were followed by a phase II trial of oral forodesine 200 mg daily. This oral formulation showed only partial activity, with a response rate of 11%, likely attributable to underdosing. Common adverse events in these trials included infection, fatigue, peripheral edema, nausea, pruritus, headache, and insomnia. Expert opinion: IV and oral formulations of forodesine have demonstrated partial activity and an acceptable safety profile in patients with refractory CTCL. A higher oral dose, or sequential therapy consisting of IV forodesine followed by maintenance oral forodesine, may be more effective. With proper dosing, forodesine may emerge as a safe and effective treatment for refractory CTCL.

Immucillins ImmA and ImmH Are Effective and Non-toxic in the Treatment of Experimental Visceral Leishmaniasis.

BACKGROUND: Immucillins ImmA (IA), ImmH (IH) and SerMe-ImmH (SMIH) are synthetic deazapurine nucleoside analogues that inhibit Leishmania (L.) infantum chagasi and Leishmania (L.) amazonensis multiplication in vitro without macrophage toxicity. Immucillins are compared to the Glucantime standard drug in the chemotherapy of Leishmania (L.) infantum chagasi infection in mice and hamsters. These agents are tested for toxicity and immune system response. METHODOLOGY/PRINCIPAL FINDINGS: BALB/c mice were infected with 107 amastigotes, treated with IA, IH, SMIH or Glucantime (2.5mg/kg/day) and monitored for clinical variables, parasite load, antibody levels and splenocyte IFN-γ, TNF-α, and IL-10 expression. Cytokines and CD4+, CD8+ and CD19+ lymphocyte frequencies were assessed in uninfected controls and in response to immucillins. Urea, creatinine, GOT and GPT levels were monitored in sera. Anti-Leishmania-specific IgG1 antibodies (anti-NH36) increased in untreated animals. IgG2a response, high levels of IFN-γ, TNF-α and lower levels of IL-10 were detected in mice treated with the immucillins and Glucantime. Immucillins permitted normal weight gain, prevented hepato-splenomegaly and cleared the parasite infection (85-89%) without renal and hepatic toxicity. Immucillins promoted 35% lower secretion of IFN-γ in uninfected controls than in infected mice. IA and IH increased the CD4+ T and CD19+ B cell frequencies. SMIH increased only the proportion of CD-19 B cells. IA and IH also cured infected hamsters with lower toxicity than Glucantime. CONCLUSIONS/SIGNIFICANCE: Immucillins IA, IH and SMIH were effective in treating leishmaniasis in mice. In hamsters, IA and IH were also effective. The highest therapeutic efficacy was obtained with IA, possibly due to its induction of a TH1 immune response. Low immucillin doses were required and showed no toxicity. Our results disclose the potential use of IA and IH in the therapy of visceral leishmaniasis.

Final results of a multicenter phase II study of the purine nucleoside phosphorylase (PNP) inhibitor forodesine in patients with advanced cutaneous T-cell lymphomas (CTCL) (Mycosis fungoides and Sézary syndrome).

BACKGROUND: Forodesine is a potent inhibitor of purine nucleoside phosphorylase (PNP) that leads to intracellular accumulation of deoxyguanosine triphosphate (dGTP) in T and B cells, resulting in apoptosis. Forodesine has demonstrated impressive antitumor activity in early phase clinical trials in cutaneous T-cell lymphoma (CTCL). PATIENTS AND METHODS: In this phase II study, patients with CTCL who had already failed three or more systemic therapies were recruited. We investigated the response rate, safety and tolerability of oral forodesine treatment in subjects with cutaneous manifestations of CTCL, stages IB, IIA, IIB, III and IVA. The safety population encompassing all stages was used for analysis of accountability, demographics and safety. The efficacy population differed from the safety population by exclusion of stage IB and IIA patients. RESULTS: All 144 patients had performance status 0-2. The median duration of CTCL from diagnosis was 53 months (5-516 months). The median number of pretreatments was 4 (range: 3-15). No complete remissions were observed. In the efficacy group of patients, 11% achieved partial remission and 50% had stable disease. The median time to response was 56 days and the median duration of response was 191 days. A total of 96% of all treated patients reported one or more adverse events (AEs) and 33% reported a serious AE. The majority of AEs were classified as mild or moderate in severity. The most commonly reported AEs (>10%) were peripheral edema, fatigue, insomnia, pruritus, diarrhea, headache and nausea. Overall eight patients died during the study: five due to sepsis and infections, one due to a second malignancy (esophageal cancer), one due to disease progression and one due to liver failure. CONCLUSION: Oral forodesine at a dose of 200 mg daily is feasible and shows partial efficacy in this highly selected CTCL population and some durable responses.

Phase I study of BCX1777 (forodesine) in patients with relapsed or refractory peripheral T/natural killer-cell malignancies.

BCX1777 (forodesine), a novel purine nucleoside phosphorylase inhibitor, induces apoptosis, mainly in T cells. To evaluate the safety, tolerability, and pharmacokinetics of BCX1777, we conducted a phase I study in patients with relapsed or refractory peripheral T/natural killer-cell malignancies. Eligible patients had relapsed or refractory peripheral T/natural killer-cell malignancies without any major organ dysfunction. BCX1777 was administered orally once daily (dose escalation: 100, 200, and 300 mg) until disease progression requiring new therapy or unacceptable adverse events occurred. A total of 13 patients were enrolled and treated in three dose cohorts (100 mg/day, five patients; 200 mg/day, three patients; 300 mg/day, five patients). Although none of the patients developed dose-limiting toxicities, further dose escalation was not performed based on data from overseas. Therefore, the maximum tolerated dose was not determined. Adverse events of grade 3 or greater (≥2 patients) included lymphopenia (62%), anemia (15%), leukopenia (8%), and pyrexia (8%). Plasma pharmacokinetics parameter of BCX1777 (area under the plasma concentration-time curve) at day 1 in each cohort was 1948 ± 884, 4608 ± 1030, and 4596 ± 939 ng•h/mL, respectively. Disease control was achieved in approximately half of patients. One patient with anaplastic large cell lymphoma, which was negative for anaplastic lymphoma kinase, achieved a complete response, and two patients with cutaneous T-cell lymphoma achieved partial responses. BCX1777 was well tolerated at doses up to 300 mg once daily and showed preliminary evidence of activity in relapsed or refractory peripheral T/natural killer-cell malignancies, warranting further investigation.

A spicamycin derivative (KRN5500) provides neuropathic pain relief in patients with advanced cancer: a placebo-controlled, proof-of-concept trial.

CONTEXT: Neuropathic pain in patients with cancer can be difficult to treat effectively. OBJECTIVES: The purpose of the study was to determine safety and efficacy of KRN5500, a novel, spicamycin-derived, nonopioid analgesic agent, in patients with advanced cancer and neuropathic pain of any etiology. METHODS: The study was a Phase 2a, multicenter, double-blind, placebo-controlled, dose escalation clinical trial. Patients with refractory neuropathic pain and advanced cancer were randomly assigned 2:1 to receive a maximum of eight single escalating doses of KRN5500 or placebo, ranging from 0.6 to 2.2 mg/m(2). The primary objective was safety and tolerability. The secondary objective was efficacy, measured by change in average pain intensity on a 0-10 numeric rating scale administered one week after the patient's final dose. RESULTS: Nineteen patients received treatment (KRN5500 n=12; placebo n=7). The most frequently reported adverse events were gastrointestinal symptoms, which were more frequent and severe with KRN5500 than placebo; two (17%) KRN5500 patients discontinued the study because of nausea and vomiting. At study endpoint, KRN5500 exhibited a significant median decrease in pain intensity from baseline of 24% compared with 0% for placebo (P=0.03). The median for largest weekly reduction in target pain intensity was 29.5% for KRN5500 and 0% for placebo patients (P=0.02). CONCLUSION: This proof-of-concept study for KRN5500 in patients with advanced cancer and any type of neuropathic pain found gastrointestinal adverse events to be the predominant safety concern. The results also provided the first indication of clinical and statistical efficacy in reducing pain intensity.

[Oral cladribine for relapsing-remitting multiple sclerosis: another purine analogue or a genuine therapeutic innovation?]. / Cladribin-Tabletten bei schubförmiger Multipler Sklerose: Noch ein Purinanalogon oder echte therapeutische Innovation?

Azathioprine (AZA) is a purine analogue which has been used in the treatment of multiple sclerosis (MS) for over 30 years. After the approval of immunomodulatory drugs, such as recombinant interferon beta and glatiramer acetate, AZA now only plays a minor role in MS therapy. The results of a recently published phase III trial (CLARITY trial) involving an oral formulation of the purine analogue cladribine, a substance which is at least structurally related to AZA, may soon lead to the approval of cladribine tablets as a new oral MS therapeutic. The following overview provides a comparison of the mode of action, side-effect profile and data currently available on AZA and cladribine in the treatment of MS.

Activation of adenosine 2A receptors preserves structure and function of podocytes.

Adenosine 2A receptor (A(2A)R) activation was recently shown to be renoprotective in diabetic nephropathy. A(2A)R are found in glomeruli and have been shown to associate with the podocyte cytoskeletal protein alpha-actinin-4, but the effect of their activation on podocyte structure and function is unknown. Podocyte injury was induced in C57BL/6 mice with puromycin aminonucleoside, and the selective A(2A)R agonist ATL313 was found to attenuate the resulting albuminuria and foot process fusion. The selective A(2A)R antagonist ZM241385 reversed the effects of ATL313. In vitro, A(2A)R mRNA and protein were expressed in a conditionally immortalized podocyte cell line, and A(2A)R-like immunoreactivity co-localized with the actin cytoskeleton. Treatment with ATL313 also blocked the increased podocyte permeability to albumin and disruption of the actin cytoskeleton that accompanied puromycin aminonucleoside-induced injury in vitro. ATL313 was ineffective, however, in the presence of the A(2A)R antagonist and in A(2A)R-deficient podocytes. It was concluded that A(2A)R activation reduces glomerular proteinuria, at least in part, by preserving the normal structure of podocyte foot processes, slit diaphragms, and actin cytoskeleton.

Nelarabine: a novel purine antimetabolite antineoplastic agent.

BACKGROUND: Nelarabine was approved by the US Food and Drug Administration (FDA) in October 2005 for the treatment of T-cell acute lymphoblastic leukemia (T-ALL) and T-cell lymphoblastic lymphoma (T-LBL) that has not responded to or has relapsed after treatment with at least 2 chemotherapy regimens. OBJECTIVES: This article reviews the pharmacology, mechanism of action, and pharmacokinetic and pharmacodynamic properties of nelarabine. Also reviewed are nelarabine's clinical efficacy in T-ALL, T-LBL, and other hematologic malignancies; its toxicity profile, dosage, and administration; and areas of ongoing and future research. METHODS: Relevant literature was identified through searches of MEDLINE (1966-April 2007), International Pharmaceutical Abstracts (1970-April 2007), and the American Society of Hematology database (2003-2006) using the terms nelarabine, Arranon, 506U78, and 2-amino-6-methoxypurine arabinoside. The reference lists of the identified articles were searched for additional sources. Product information obtained from the manufacturer of nelarabine was consulted, as were the FDA reviews of nelarabine. All identified publications were considered, and those meeting the objectives of this review were included. RESULTS: Nelarabine, a soluble prodrug of 9-beta-D- arabinofuranosylguanine (ara-G), is a novel purine antimetabolite antineoplastic that preferentially accumulates in T-cells. Ara-G is rapidly phosphorylated in T-cells to ara-G triphosphate (ara-GTP), which exerts cytotoxic effects. Renal elimination of ara-G is decreased in patients with mild to moderate renal impairment; however, no dose adjustment is recommended. Accumulation of ara-GTP occurs in T-cells in a dose-dependent manner, leading to preferential cytotoxicity. Nelarabine has activity in T-cell malignancies, as evaluated in 2 Phase I and 5 Phase II studies. It received accelerated approval from the FDA based on the resuits of 2 Phase II trials, one in pediatric patients (PGAA 2001) and the other in adults (CALGB 19801). In PGAA 2001, patients with T-ALL in first relapse (n = 33) had an objective response rate of 55% (16 with a complete response [CR] and 2 with a partial response [PR]), and those with T-ALL in second relapse (n = 30) had an objective response rate of 27% (7 CR and 1 PR). Among patients with central nervous system-positive T-ALL or T-cell non-Hodgkins lymphoma (T-NHL) (n = 21), 33% had an objective response (5 CR and 2 PR); among patients with T-ALL or T-NHL with extramedullary relapse (n = 22), 14% had a PR. CALGB 19801 included 39 adult patients with T-cell malignancies, of whom 7 (18%) had a CR and an additional 2 (5%) had a CR without full hematologic recovery. The recommended dose of nelarabine in adults is 1500 mg/m(2) IV given over 2 hours on days 1, 3, and 5, repeated every 21 days; the recommended dose in pediatric patients is 650 mg/m(2) IV given over 1 hour for 5 consecutive days, repeated every 21 days. Dose-limiting toxicities observed in the Phase I and II trials included central and peripheral neurotoxicity. Symptoms of central neurotoxicity included somnolence, seizures, dizziness, confusion, and ataxia; symptoms of peripheral neurotoxicity included paresthesias, pain in the extremities, and peripheral neuropathy. CONCLUSIONS: Nelarabine is indicated for the treatment of T-ALL and T-LBL that has not responded to or has relapsed after treatment with at least 2 chemotherapy regimens. Objective response rates in Phase II clinical trials of nelarabine have ranged from 11% to 60%. Use of nelarabine is limited by potentially severe neurotoxicity.

[Spicamycin derivative].

Spicamycin, induced differentiation of human myeloid leukemia cells (HL-60), is found to show a potent antitumor activity by inhibiting protein synthesis. Among the various semisynthetic derivatives of spicamycin, KRN5500 showed a marked efficacy in human tumor xenograft model. KRN5500 itself has, however, only minor inhibitory effects on protein synthesis in cell free system. A metabolite, SAN-Gly, which is thought to be generated through metabolism of KRN5500 by a cytosomal enzyme, exhibited a marked inhibitory effect. KRN5500 is expected to be useful even for treatment of refractory solid tumors because of its unique antitumor mechanism. A phase I clinical trials underwent at the National Cancer Center Hospital in Tokyo and at the National Cancer Institute in the USA. Unfortunately, the drug toxicities in addition to grade 4 pulmonary disorders were occurred, partly caused by the organic solvents and chemical essential for its dissolution. To overcome such a severe adverse effects, we conducted to examine whether incorporation of KRN5500 into polymeric micelles (KRN/m) could reduce a pulmonary disorder using a bleomycin (BLM)-induced lung injury rat model or exhibit antitumor activity similar to KRN5500. In conclusion, this study demonstrated that KRN/m is superior to KRN5500 because the pulmonary toxicity was reduced and the potent antitumor activity of KRN5500 was retained after the incorporation of KRN5500 into micelles. We think that these results justify a clinical phase I trial of KRN/m.

Transformation and progression of Waldenström's macroglobulinemia following cladribine therapy in two cases: natural evolution or iatrogenic causation?

We report two cases of Waldenström's macroglobulinemia with an unusual aggressive transformation following treatment with cladribine (2-chlorodeoxyadenosine, 2-CdA), a purine analogue. The first patient developed transformation to a diffuse large-cell non-Hodgkin lymphoma, while the second developed extensive extramedullary involvement. Both patients displayed rapid progression following transformation and were refractory to chemotherapy. Both patients were pretreated with multiple courses of prednisone and chlorambucil, and transformation occurred shortly after therapy with cladribine. We propose that immune suppression from alkylating agents and purine analogues may have contributed to the unusual progression, resulting in a dismal outcome.

Purine nucleoside analogues for the treatment of hematological malignancies: pharmacology and clinical applications.

The purine nucleoside analogues (PNAs), fludarabine (FA), 2-CdA (2-chlorodeoxyadenosine, 2-CdA) and pentostatin (2'-deoxycoformycin, DCF) represent a group of cytotoxic agents with high activity in lymphoid and myeloid malignancies. PNAs share similar chemical structure and mechanism of action. Several mechanisms could be responsible for their cytotoxicity both in proliferating and quiescent cells, such as inhibition of DNA synthesis, inhibition of DNA repair and accumulation of DNA strand breaks. Induction of apoptosis through the mitochondrial pathway, direct binding to apoptosome or modulation of p53 expression all lead to apoptosis, which is the main end-point of PNA action. However, individual PNAs exhibit significant differences, especially in their interaction with enzymes involved in adenosine and deoxyadenosine metabolism. Synergistic interactions between PNAs and other cytotoxic agents (alkylating agents, anthracycline antitumor antibiotics, cytarabine, monoclonal antibodies) have been demonstrated in both preclinical and clinical studies. PNAs are highly effective in chronic lymphoid leukemias and low grade B- and T-cell non-Hodgkin's lymphomas, including Waldenström's macroglobulinemia. DCF and 2-CdA are currently the drugs of choice in hairy cell leukemia. Moreover, clinical studies have confirmed the efficacy of PNAs alone or in combination protocols in the treatment of acute myeloid leukemia and myelodysplastic syndromes. Finally, PNAs, especially FA, play an important role in non-myeloablative conditioning regimens for allogenic stem cell transplantation in high-risk patients. The toxicity profiles of PNAs are similar for all agents and consist mainly of dose-limiting myelotoxicity and prolonged immunosuppression. Three other compounds: clofarabine, nelarabine and immucillin-H are currently being evaluated clinically.

New nucleoside analogs in the treatment of hematological disorders.

Cytotoxic nucleoside analogs have a broad clinical use. They were among the first chemotherapeutic agents used in the treatment of malignant diseases. The anticancer nucleosides include analogs of physiologic pyrimidine and purine nucleosides. They are used in oncology in the treatment of both, solid tumors and hematological malignancies. These agents have many intracellular targets, e.g. they act as antimetabolites, competing with natural nucleosides during DNA or RNA synthesis and as inhibitors of key cell enzymes. Understanding of the mechanisms of action of these compounds and synthesis of new analogs provides the possibility to further expand the spectrum of their clinical use and enhance their antitumor activity. In this paper we describe mechanisms of action and possible clinical use in the treatment of hematological malignancies of these nucleoside analogs, which are now in different stages of clinical trials, namely tezacitabine, troxacitabine, clofarabine, nelarabine, decitabine, CNDAC and ECyD.

Drug-induced pulmonary infection.

Immunosuppressive drugs lead to an enhanced risk for infection. The impact of these drugs on the immune system can be broad (eg, corticosteroids) or targeted (eg, rituximab). Infections can have serious consequences, particularly if there is a delay in diagnosis. It is hoped that a knowledge of the type of immune defects that are induced by these drugs and the specific infections that have been reported will guide clinicians in the appropriate use of prophylactic regimens and diagnostic considerations in the event of pneumonia.

Phase I and pharmacokinetic study of KRN5500, a spicamycin derivative, for patients with advanced solid tumors.

BACKGROUND: KRN5500, a novel spicamycin derivative, shows an inhibitory effect on protein synthesis. This phase I study was aimed at investigating the toxicity, maximum tolerated dose (MTD) and pharmacokinetics of this compound. PATIENTS AND METHODS: Patients with solid tumors not amenable to standard forms of treatment were eligible. KRN5500 was administered as a 2 h intravenous infusion every 4 weeks at doses of 3, 6, 10, 15 and 21 mg/m(2). Pharmacokinetic evaluation was performed at the first cycle. RESULTS: Eighteen patients with advanced solid tumors were enrolled. A total of 26 cycles of KRN5500 were administered. The major toxicities were nausea, vomiting, diarrhea, fatigue and a mild reversible prolongation of prothrombin time. Grade 4 pulmonary toxicity (interstitial pneumonitis) was observed in one patient at a dose level of 15 mg/m(2). Severe fatigue was observed in one patient at a dose level of 21 mg/m(2) and the duration of fatigue tended to increase with the dose of KRN5500. Nausea and vomiting were frequently observed and became prolonged with increasing dose of KRN5500. These toxicity profiles were identified as unacceptable and further dose escalation above 21 mg/m(2) was withheld. The MTD was therefore determined as 21 mg/m(2). The peak plasma concentration and the area under the concentration-time curve of KRN5500 increased proportionally to the dose, suggesting linear pharmacokinetics. No objective antitumor response was observed. CONCLUSION: KRN5500, a structurally novel protein synthesis inhibitor, warrants further investigation to overcome these toxicity profiles and improve its efficacy.

A phase I clinical trial of spicamycin derivative KRN5500 (NSC 650426) using a phase I accelerated titration "2B" design.

The spicamycin derivative KRN5500 was considered as a potential anti-cancer agent based on in vitro and preclinical studies. A Phase I study involving 24 cancer patients in whom tumors were refractory to all other conventional therapies was conducted to determine the dose limiting toxicity, maximum tolerated dose, effectiveness, and pharmacokinetic parameters of this drug administered by 1-h IV infusion daily for five consecutive days every 3 weeks. Using an accelerated dose titration strategy, 8.4 mg/m2/d x 5 days was the maximum administered dose. Severe gastrointestinal and hepatic toxicities were observed at doses at or above 4.3 mg/m2/d x 5. The recommended Phase II dose i s 4.3mg/m2/d x 5. The distribution of KRN5500 followed a two-compartment model, and clearance did not decrease significantly over the dose range 0.8-8.4 mg/m2/d x 5. No significant correlation was observed between plasma levels and toxicity. No tumor responses were observed among the 14 patients evaluable for response.

Complications in the treatment of CLL with purine analogues.

The main complications from treatment of CLL with purine analogues are bone marrow depression resulting in cytopenia, immune suppression leading to infections, particularly opportunistic infections, and immune deregulation with autoimmune hemolytic anemia as the most common feature. These complications usually delay or even prevent further treatment, thus leading to a reduced clinical response. In order to achieve the greatest benefit from purine analogue therapy, it is important to be able to predict the risk of such complications, and to use a treatment strategy with optimal dose intensity, by prophylactic supportive care when required, and by early detection and adequate management of complications.

Neurotoxicity of purine analogs: a review.

PURPOSE: The purine analogs, fludarabine, cladribine, and pentostatin, are active against a broad spectrum of indolent lymphoid malignancies. They also have similar toxicities, including myelosuppression, immunosuppression, and sporadic neurotoxicity. This review compares the spectrum of neurotoxicity of each of these agents. Now that these drugs are commercially available and are being widely used, physicians should be aware of potentially serious side effects that may be encountered. METHODS: The literature was searched using MedLine and Cancerline, as well as the bibliographies of published reports through the fall of 1993. In addition, case records from National Cancer Institute (NCI) Group C protocols were reviewed for fludarabine in chronic lymphocytic leukemia (CLL), and cladribine and pentostatin in hairy cell leukemia (HCL), as well as adverse drug reactions reported to the NCI from January 1980 through September 1993. RESULTS: At higher than recommended doses, life-threatening and fatal neurotoxicity were encountered with all three drugs. At the recommended doses, each agent induced neurotoxicity in approximately 15% of patients, mostly mild and reversible. However, severe neurologic complications were reported; these were occasionally delayed, sometimes fatal, but often at least partially reversible. CONCLUSION: The doses of these three agents should not be increased above the recommended levels. Development of moderate or worse neurotoxicity should result in discontinuation of that drug.