Phase I Evaluation of a 40-kDa Branched-Chain Long-Acting Pegylated IFN-α-2a With and Without Cytarabine in Patients with Chronic Myelogenous Leukemia Free

Purpose: Pegasys (PEG-IFN) is a modified form of recombinant human IFN-α-2a in which IFN-α is attached to a branched methoxypolyethylene glycol (PEG) moiety of large molecular weight (40 kDa). Such molecular modification results in sustained absorption after s.c. drug administration and a prolonged half-life. A phase I study of PEG-IFN was conducted in patients with chronic myelogenous leukemia (CML) who were previously treated with IFN-α to evaluate the effect of sustained exposure to IFN on patients with CML.

Experimental Design: Twenty-seven patients with long-term or IFN-refractory CML were enrolled in cohorts of three or six patients. PEG-IFN was given once weekly by s.c. injections starting at a dose of 270 μg/wk to a maximum dose of 630 μg/wk. Sixteen additional patients were treated with escalating doses of PEG-IFN ranging from 450 to 540 μg/wk in combination with two different schedules of low-dose cytarabine (1-β-d-arabinofuranosylcytosine, ara-C). Serial venous blood samples were collected to evaluate the pharmacokinetic and pharmacodynamic characteristics of PEG-IFN in these patients.

Results: The dose-limiting toxicity (DLT) as defined by the protocol was not achieved at the highest dose tested of 630 μg/wk. With the addition of ara-C, the DLT was reached at 540 μg/wk. The safety profile was similar to that of unmodified IFNs. Of 27 patients treated with PEG-IFN, 14 (52%) achieved or maintained a complete hematologic response and three (11%) achieved a complete cytogenetic response. Among 16 patients treated with the combination of PEG-IFN and ara-C, 11 (69%) achieved or maintained complete hematologic remission and two (13%) achieved complete cytogenetic remission. The mean serum peak concentration (C_max) of PEG-IFN increased from 9.4 to 28 ng/mL as the dose increased from 270 to 450 μg/wk, with no further increases in C_max at higher dose levels. Serum concentration reached peak value starting about 48 hours after drug administration and was maintained at close to peak value throughout the dosing interval. The mean ± SD area under the serum concentration-time curve (AUC) calculated after the first dose also increased from 1,022 ± 694 to 3,343 ± 2,728 ng hour/mL as dose was increased from 270 to 450 μg/wk, showing a dose-related increase in systemic exposure of PEG-IFN. As with C_max, the AUC did not increase at higher dose levels. The maximum induction (E_max) of neopterin, the surrogate marker of the pharmacodynamic activity of PEG-IFN, increased from 120% to 361% over baseline values as the dose was increased from 270 to 540 μg/wk. On the once-weekly multiple dosing schedule, both the PEG-IFN and neopterin concentration seemed to reach steady state by week 5 and the steady-state values were maintained with chronic dosing over 6 months.

Conclusion: Pegasys provided a significant advantage over standard IFN-α by enabling once-weekly dosing while maintaining acceptable safety, tolerability, and activity profiles. This branched 40-kDa PEG-IFN was well tolerated both as a monotherapy as well as in combination with ara-C. Demonstration of its sustained exposure, pharmacodynamic activity, hematologic response, and evidence of cytogenetic response in several patients in this limited study with either IFN-refractory or INF-intolerant patients provides a promise for further investigation in combination with new agents like imatinib.

Chronic myelogenous leukemia (CML) is a clonal myeloproliferative disorder of a pluripotent hematopoietic stem cell with a specific cytogenetic abnormality, the Philadelphia (Ph+) chromosome. The Ph+ abnormality results from a translocation between the long arms of chromosomes 9 and 22. This produces the BCR-ABL chimeric gene that expresses an abnormal fusion protein with altered tyrosine kinase activity (1, 2). Until recently, allogeneic bone marrow transplantation and IFN-α were the mainstays of therapy for CML (1, 2). However, imatinib mesylate (Gleevec, STI571), which was first studied in 1998, is currently the first line of treatment for the majority of patients with CML (2–7). Nevertheless, IFN-α remains an important therapy for CML because of its ability to induce sustained complete cytogenetic responses and to sustain event-free survival in a small percentage of patients (8). Furthermore, its unique modes of action that are distinct from those of imatinib provide the basis to explore the role of IFN-α in combination with imatinib, or in imatinib-resistant CML (9, 10). Virtually all patients receiving IFN-α therapy experience constitutional and neurologic side effects and discontinuation of treatment due to toxicity is necessary for 5% to 25% of patients. A large proportion of patients often receives less than the target dose owing to side effects. The short half-life of IFN-α requires frequent s.c. administrations (daily to thrice weekly), and most of the acute adverse events are believed to be related to high peak IFN levels. Thus, there is a great need for development of a modified IFN-α with improved pharmacokinetic properties that allow for less frequent drug administration and better tolerance.

PEG-IFN is synthesized by covalently attaching a branched methoxypolyethylene glycol (PEG) molecule to IFN-α-2a. The branched PEG reagent (11) consists of two monomethoxy PEG chains, each with an average molecular weight of ∼20,000 Da, linked to a lysine residue of IFN via urethane bonds. PEG-IFN retains in vitro biological activities characteristic of IFN-α against tumor cell lines. In contrast to IFN-α, results of early preclinical evaluation showed that serum concentrations of PEG-IFN were sustained in both rat and monkeys following s.c. injections. The pharmacodynamic profile of PEG-IFN evaluated by using 2′,5′-oligoadenylate synthetase activity in the serum showed similarly sustained levels.⁴

Although results of in vitro antiproliferative assays often showed less specific activity for PEG-IFN compared with IFN-α, the in vivo activity increased significantly because its altered pharmacokinetics provided sustained delivery of PEG-IFN during the dosing interval (12). For example, although PEG-IFN had significantly less activity in an in vitro antiproliferative assay against A-498 cells, once-weekly administration of PEG-IFN to A-498 tumor-bearing nude mice resulted in substantially greater antitumor activity than IFN-α thrice a week, where the total weekly dose of IFN protein was identical.⁵ Similarly, PEG-IFN also showed antitumor activity superior to that seen with IFN-α when given in nude mice bearing human renal cell carcinoma tumor xenografts. Two phase III randomized trials in patients with chronic hepatitis C comparing 180 μg PEG-IFN once weekly to 3 μg MIU IFN thrice a week showed superior efficacy of the pegylated form, resulting in sustained viral clearance of 39% versus 19% (P = 0.001) in patients without cirrhosis (13) and 30% versus 8% (P = 0.001) in patients with cirrhosis (14). A phase I study was conducted in 27 patients with renal cell cancer, with dose escalation from 180 to 540 μg/wk. The maximum tolerated dose was 540 μg/wk and five (19%) patients achieved a partial tumor response (15). Cytarabine had been studied by several groups and increased the response to standard IFNs in CML (16, 17). The protocol was therefore later amended to include several dose levels of PEG-IFN in combination with cytarabine. Herein, we report the results of a phase I trial of PEG-IFN alone or in combination with cytarabine, in patients with CML. The safety, tolerability, pharmacokinetic and pharmacodynamic properties, as well as the preliminary clinical activity of this novel 40-kDa pegylated IFN-α-2a are described.

Patients. This phase I trial was approved by the Institutional Review Board at The University of Texas M.D. Anderson Cancer Center and was conducted between October 1998 and November 2001. All patients enrolled in the trial were over 18 years of age and had a confirmed diagnosis of chronic phase CML (Ph+ or bcr-abl+). Additional requirements included a diagnosis of CML of >6 months duration and resistance to IFN-α therapy as defined by at least one of the following conditions: (a) loss of cytogenetic response (Ph+ increases by ≥30%), (b) a best response of minor or no cytogenetic response (Ph+, ≥35%), (c) hematologic response in the absence of a cytogenetic response, or (d) hematologic resistance defined by persistent elevation of WBC >20 × 10⁹/L despite treatment with IFN-α. Patients were excluded from the trial if they had accelerated or blastic phase CML as previously defined or if they had been screened for and expected to receive an allogeneic bone marrow transplant within the next 3 months. Patients had to have a Karnofsky performance status score of ≥80%. Patients deemed to be IFN-α intolerant were allowed to be included in the study.

Study design and definitions. This was an open label ascending dose trial with objectives to determine the maximum tolerated dose (MTD) and study the safety and efficacy of PEG-IFN. The protocol was later amended to include the same analyses of the combination of PEG-IFN with cytarabine, given by either of two different dosing regimens, 10 mg daily or 20 mg/M², 10 days/mo. PEG-IFN (Hoffmann-La Roche, Inc., Nutley, NJ) was supplied as a ready-to-use solution in single-dose 2-mL glass vials containing 180, 270, 360, 450, 540, or 630 μg/mL solution. PEG-IFN was given by s.c. injection once weekly for 48 weeks. A patient continued an assigned dose for the duration of treatment unless adverse events necessitated a decrease in dose, whereas the dose of patients who tolerated treatment without cytogenetic response was escalated to the next higher dose level. Patients who showed evidence of response at the end of 48 weeks were allowed to continue treatment for an additional year or until disease progression.

Patients in this open-label study were dosed in groups of three to six, starting at 270 μg/wk and increasing in 90 μg increments (270, 360, 450, 540, and 630 μg/wk). Dose escalation was permitted depending on the dose-limiting toxicities (DLT) encountered through day 28 of the first cycle. If no DLTs were observed in the first 28 days of treatment, patients were treated at the next higher dose level. If two or more patients had DLT, patient accrual continued at the previous dose level to define MTD. DLT was defined in the protocol as (a) grade ≥3 nonhematologic toxicity except for fever, chills, and flu-like symptoms; (2) grade ≥3 thrombocytopenia; (c) grade 4 neutropenia; or (d) according to the clinical judgment of the investigator in consultation with the sponsor (Hoffmann-La Roche). Following amendment of the protocol to include cytarabine, the highest tolerated doses of PEG-IFN were tested in a dose-ascending manner with cytarabine: either high dose for 10 days/mo or daily low dose. The dose levels tested were PEG-IFN 450 μg weekly + 1-β-d-arabinofuranosylcytosine (ara-C) 10 mg daily, PEG-IFN 540 μg weekly + ara-C 10 mg daily, and PEG-IFN 540 μg weekly + ara-C 20 mg/M² 10 days/mo. Treatment with cytarabine was to be limited to 6 months' duration for reasons of safety, and treatment with cytarabine was to be discontinued if the patient's granulocyte count dropped below 1.0 × 10⁹/L.

Dose modification was recommended for patients who experienced grade ≥3 constitutional toxicity (except for fever, chills, and flu-like symptoms), grade 4 fever, chills, and flu-like symptoms; grade ≥3 nonhematologic laboratory abnormalities; grade ≥3 thrombocytopenia, lymphopenia, or anemia; and grade 4 neutropenia. Dosing was withheld until the adverse event had resolved or was determined to be grade ≤1 in intensity (grade ≤2 for hematologic toxicity). Patients could then be redosed at the previous dose level (90 μg lower dose). If the adverse event recurred, the dose could be held and reduced once more in the same way to the next lower dose. Therapy was discontinued for those patients who could not tolerate the second dose reduction. Any patient who experienced grade 1 or 2 toxicity was maintained on the assigned dose, if possible. After amendment to include combination therapy, the dose of ara-C would be reduced to half (either 5 mg/d or 10 mg/M² for 10 days/mo, depending on regimen). If the adverse event reoccurred, the dose could be held and reduced once more in the same way (either 2.5 mg/d or 5 mg/M² for 10 days/mo).

The primary efficacy variable was defined as the rate of complete cytogenetic response. A complete cytogenetic response was defined as the absence of the Philadelphia chromosome (Ph+) in ≥20 bone marrow cells in metaphase. A partial cytogenetic response was defined as detection of Ph+ in 1% to 34% of bone marrow metaphase cells. A minor cytogenetic response was defined as 35% to 94% of cells with Ph+. The secondary efficacy variable was hematologic response rate. Time to response, duration of response, and time to disease progression were also evaluated for both cytogenetic and hematologic responses, as well as overall survival rate. No concomitant therapy or other biological agents were allowed, with the exception that a single course of hydroxyurea not in excess of 5 days was allowed if a patient had an increase in WBC count >60 × 10⁹/L. Subsequently, if the WBC count increased despite PEG-IFN therapy, the patient was withdrawn from therapy and was considered resistant to PEG-IFN treatment.

Pharmacokinetic and pharmacodynamic evaluations. Serial venous blood samples were collected at baseline, 6 to 8, 48, 72, 96, 120, 144, and 196 hours on week 1 to evaluate pharmacokinetic characteristics of PEG-IFN after the first dose of the treatment period. Additional predose (trough) samples were collected at 4-week intervals during the treatment period. Serum concentrations of neopterin were also assessed as a surrogate for evaluation of the pharmacodynamic characteristics of PEG-IFN with samples collected at same times as taken for PEG-IFN. Serum samples were stored at −20°C until assay.

Serum concentrations of antibodies to PEG-IFN were not measured in this study based on the nonmeasurable levels observed in earlier studies with cancer patients.

Measurement of serum PEG-IFN. Serum PEG-IFN concentrations were measured with a quantitative sandwich ELISA using two distinct mouse monoclonal anti-human IFN-α antibodies that recognize two different epitopes of IFN-α-2a. In a one-step immunoreaction, PEG-IFN was bound to the peroxidase-conjugated anti-Hu-IFN-α antibody. The complex binds via the capture of anti-IFN-α antibody to the multiprotein layer-covered surface of the microtiter plate. Unbound complexes were removed by washing. Tetramethylbenzidine substrate solution was then added. Conversion of tetramethylbenzidine by peroxidase to a colored product allowed the reaction to be followed spectrophotometrically, as the absorbance at 450 nm was directly correlated to the concentration of PEG-IFN in the sample. The analytic sensitivity was about 125 pg/mL of PEG-IFN in human serum. The intra-assay variation of the quality control samples was between 0.8% and 12%, whereas the interassay variation was between 4.5% and 18.3%. PEG-IFN was stable in serum over three freeze-thaw cycles. Samples stored at −20°C were stable over 5 months.

Measurement of serum neopterin. The serum concentration of neopterin was measured using a commercial RIA kit (HENNING test, BRAHMS Diagnostica GmbH, D-12064, Berlin, Germany) consisting of a single incubation step and using a double antibody phase separation technique. In brief, ¹²⁵I-neopterin tracer and the premix of the first antibody (specific to neopterin) and a second antibody (anti-sheep IgG) were pipetted into the study sample as described in the kit. After mixing thoroughly, the samples were incubated in darkness for 1 hour at between 20°C and 25°C. When the immunoreaction was complete, the tubes were centrifuged, the supernatants aspirated, and the pellets counted in a γ counter for 5 minutes. The amount of radioactivity remaining in the pellet was inversely proportional to the concentration of neopterin found in the sample. The limit of quantitation was 3 nmol/L using a 50-μL sample. The intra-assay and between-assay variabilities ranged from 1.8% to 6.5%.

Pharmacokinetic and pharmacodynamic data analysis. Serum concentration (or activity) versus time data were analyzed for PEG-IFN and neopterin by noncompartmental pharmacokinetic methods. The highest observed concentration and the corresponding sampling time were defined as C_max and t_max, respectively. The area under the concentration-versus-time curve [AUC_(0-τ)] was calculated by use of the trapezoidal rule over the dosing interval of 1 week.

Patient characteristics. Forty-three patients (25 males and 18 females) were treated on study (Table 1). The median age was 52 years (range, 24-69 years). All patients had chronic phase CML (median disease duration, 4.5 years) that had been previously treated with IFN-α; some patients had received additional CML therapy: hydroxyurea (27 patients), Ara-C (29 patients), homoharringtonine (10 patients), and others. The patients' previous responses to IFN were either primary resistance (17 patients), refractory (25 patients), or intolerance (one patient). Four patients were intolerant to IFN treatment as well as being either resistant or refractory. Before entering the study, 29 patients had progressed cytogenetically, and eight had progressed hematologically. In the six remaining cases, other reasons were provided, including intolerance.

Twenty-seven patients were treated with PEG-IFN monotherapy: three patients with 270 μg weekly, and six patients each were assigned to 360, 450, 540, and 630 μg weekly (five cohorts). Sixteen patients received combination of weekly PEG-IFN with one of two dose schedules of ara-C: PEG-IFN + ara-C: 450 μg weekly + ara-C 10 mg/d (six patients) or 540 μg weekly + ara-C 10 mg/d (six patients) or 540 μg weekly + 20 mg/M² × 10 days/mo; four patients, three cohorts). One patient assigned to the 540 μg PEG-IFN + ara-C 10 mg/d combination group inadvertently received 450 μg PEG-IFN in week 1 followed by 540 μg PEG-IFN in subsequent weeks. For analysis purposes, this patient is included in the 540 μg PEG-IFN + ara-C 10 mg/d dose group, except for analysis of the first dose pharmacokinetic/pharmacodynamic, where the actual dose of PEG-IFN (450 μg) was used (Table 2).

Treatment administered and toxicity. MTD was defined by the occurrence of DLT occurring during the first 28 days of treatment. DLT occurred within the first 28 days of therapy with the two PEG-IFN 540 μg + ara-C combinations (one patient each with grade 4 neutropenia and grade 3 skin rash/mucositis and two patients with grade 3 thrombocytopenia) but not in any monotherapy cohorts. Hence, MTD could be defined only for the combination treatment, at PEG-IFN 450 μg once weekly + ara-C 10 mg/d.

Over the full course of treatment, the overall side effect profile was mostly mild to moderate in intensity. The toxicities most frequently reported, regardless of intensity or relationship to treatment, included fatigue, nausea, headache, diarrhea, myalgia, decreased appetite, and arthralgia (Table 3). Of the 27 patients treated with PEG-IFN alone, the most commonly reported treatment-related grade 3 or 4 toxicities included fatigue (n = 11 patients), headache (n = 7 patients), nausea (n = 6 patients), diarrhea, (n = 6 patients), decreased appetite (n = 6 patients), and myalgia (n = 6 patients). The most common grade 3 to 4 toxicities reported by patients treated with combined therapy were fatigue (n = 7 patients), headache (n = 3 patients), and decreased appetite (n = 3 patients). Mild to moderate alanine transferase (ALT) elevations occurred in some patients at all dose levels, and grade 2 to 3 ALT resulted in dose reduction in five patients. Grade 2 to 3 ALT toxicity was reported for three patients on monotherapy at week 13 and for three patients on combination treatment between weeks 3 and 13. Toxicity was managed by withholding therapy (for an average of 6 weeks) until ALT levels normalized, then drug was restarted at a lower dose. When dose modification was instituted earlier, ALT was maintained at a lower level.

Over the entire study period, 25 of 27 (93%) patients in the monotherapy group required dose reductions. The median time from start of treatment to dose modification was <3 months for 6 of 27 (22%) patients. Three of these six (50%) patients were in the high dose group of 630 μg. Thirteen reductions occurred between 3 and 6 months, 12 dose reductions were between 6 and 9 months, and 15 reductions between 9 and 12 months. Because a patient could have multiple dose reductions, the sum of dose reductions exceeds the number of patients on the study. After all dose reductions were made, the mean PEG-IFN doses received were 214.2, 386.7, 360.9, 487.3, and 401.3 μg, respectively, by dose group.

Of the 16 patients in the combination therapy PEG-IFN and ara-C group, 15 (94%) patients required dose adjustments over the entire study period. Eleven dose reductions occurred in the first 3 months. All of the patients who received 540 μg/wk of PEG-IFN + ara-C required dose reductions and five of six patients who received the lower dose of 450 μg/wk required a reduction. After all dose reductions were made, the mean PEG-IFN doses received were 350.1, 275.8, and 394.9 μg, respectively, by dose group.

Thirteen patients were removed from the study because of adverse events (Table 4). Twelve of these were due to drug-related side effects and one patient developed a lymphoma as a second malignancy. Of the 27 patients in the monotherapy groups, seven withdrew for reasons including weakness, fatty liver, increased bilirubin, thrombocytopenia, stomatitis, abnormal hepatic function, and lymphoma. Of the 16 patients in the combination treatment groups, six withdrew due to mucosal inflammation/pruritic rash, fatigue (two patients), abnormal hepatic function, limb pain, and aggravated peripheral neuropathy.

Of 43 patients treated in the study, 23 (14 monotherapy and nine combination) completed the full 1 year of treatment (details provided in Table 4).

Efficacy. Fifteen of the 27 (56%) patients treated with Pegasys monotherapy maintained or achieved complete hematologic response. Twelve of the 16 patients treated with the combination therapy with ara-C exhibited a similar response (Table 5A). Complete cytogenetic responses were observed in 6 of the 43 (14%) patients, including 3 of 27 (11%) on the monotherapy and 3 of 16 (19%) on the combination therapy (Table 5B). Four patients also had partial cytogenetic responses. Overall incidence of major (complete and partial) cytogenetic response was 10 of 43 (23%). For the most part, patients who had previously responded to IFN were the ones who responded to Pegasys, but there were three cases of new or improved hematologic response and three cases of new or improved cytogenetic response. Four of the six (67%) patients achieved complete cytogenetic responses that lasted for 6 to 43 months. One maintained major cytogenetic response (1 of 20 Ph′) but was taken off study and placed on imatinib. Four of these patients lost response because of treatment discontinuation due to toxicities. One was taken off study after 12 months of complete cytogenetic remission to be placed on imatinib. Four patients had partial cytogenetic remission lasting for 6 to >16 months (when patient was taken off study with ongoing partial remission to be placed on imatinib).

Pharmacokinetic properties: Monotherapy. The pharmacokinetic characteristics of PEG-IFN after the first dose are summarized in Table 2. Serum concentrations of PEG-IFN increased rapidly after s.c. administration, reaching close to peak concentration at about 48 hours after drug administration for most groups. Serum concentrations were subsequently sustained at close to peak levels throughout the rest of the dosing interval (168 hours). The C_max after the first dose increased as dose increased for the three low-dose monotherapy groups: from 9 to 28 ng/mL at 270 to 450 μg dose. However, this linear relationship broke down at the higher doses as C_max did not seem to increase when the dose was increased further to 540 (26 ng/mL) and 630 μg/wk (27 ng/mL), suggesting potential nonlinearity in systemic availability of PEG-IFN at doses higher than 450 μg/wk doses. There was considerable interpatient variability within dose groups, and sample size was relatively small.

The area under the serum concentration-time curve showed similar behavior as C_max (Table 2). It increased in a linear manner from 1,022 ± 694 ng hour/mL at a dose of 270 μg to 3,343 ± 2,728 ng hour/mL at a dose of 450 μg of PEG-IFN given as weekly monotherapy, with essentially no further increase in AUC at higher doses. Because of the sustained serum concentration during the dosing interval of 7 days, the terminal half-life of serum PEG-IFN could not be evaluated in this study. Following once-weekly doses, steady-state PEG-IFN levels were attained after ∼5 to 9 weeks, with PEG-IFN concentrations increasing about 3- to 5-fold after initial dosing. After steady state was achieved, the predose serum concentration remained stable (Fig. 1A).

Combination therapy with 1-β-d-arabinofuranosylcytosine. Ara-C given either at a dose of 10 mg/d (two cohorts) or 20 mg/m² for 10 days/mo did not result in any significant changes in PEG-IFN serum concentration compared with monotherapy (Table 2). For example, mean ± SD serum AUC for PEG-IFN at 450 μg dose with and without ara-C (10 mg) were 3,343 ± 2,728 and 3,003 ± 2,121 ng/mL, respectively.

Pharmacodynamic properties: Monotherapy. Significant induction of neopterin production was observed at all doses studied (Table 2). Maximum induction was mostly observed 48 to 96 hours after administration of the first dose of PEG-IFN. Neopterin concentrations subsequently remained above baseline throughout the entire dosing interval. Maximum induction (percentage change from baseline) increased with increasing doses of 270 and 360 μg but did not increase any further at higher doses of 450, 540, or 630 μg/wk (Table 2). As with PEG-IFN monotherapy, there was also large interpatient variability in neopterin induction within a dose group and number of patients in each group was relatively small (n = 3-7). Similar to PEG-IFN, neopterin induction was maintained at close to its maximum level at the end of dosing interval (168 hours). Predose induction of neopterin was, therefore, maintained throughout chronic dosing, although multiple dose accumulation was less apparent for neopterin than for PEG-IFN levels (Fig. 1B).

Combination therapy with 1-β-d-arabinofuranosylcytosine. Coadministration of ara-C did not seem to have an appreciable effect on induction of neopterin as compared with PEG-IFN monotherapy regardless of the dose of ara-C or PEG-IFN (Table 2).

Dose-effect relationship. In spite of large interpatient variability, the average maximum drop in WBC count seemed to increase with dose, up to a dose of 450 μg/wk; further increases in the PEG-IFN dose did not seem to cause any further decrease in WBC count. The mean (SE) percent drop in WBC at doses of 270, 360, 450, 540, and 630 μg were 32.4 (16.7), 51.6 (10.1), 64.3 (14.3), 62.9 (13.4), and 59.6 (9.5), respectively. Administration of ara-C with PEG-IFN was not associated with any further decrease in WBC. As with the steady-state PEG-IFN levels, it required about 5 to 9 weeks to achieve a maximum percentage decrease in WBC counts. PEG-IFN caused (mostly grade 1) elevations of serum transaminases in most of the patients irrespective of dose given. There was no predictable relationship between predose (trough) PEG-IFN level and the severity of ALT increase in terms of National Cancer Institute Common Toxicity Criteria grade. Median trough PEG-IFN values at grades 0 to 3 of ALT were 23.2, 47.8, 36.5, and 53.9 ng/mL, respectively.

Of 27 patients treated with PEG-IFN monotherapy at 270 to 630 μg/wk, three had complete cytogenetic responses, one had a partial response, and five had minor response. The remaining 18 patients did not show any clinically significant cytogenetic response. Similarly, of 16 patients treated with combination PEG-IFN and ara-C, there were three complete cytogenetic responses, three partial responses, and three minor responses. The remaining seven patients did not show any cytogenetic response. There were no predictable cytogenetic responses in patients who showed complete or partial hematologic response. The onset of cytogenetic response was 36 to 48 weeks in contrast to almost immediate hematologic response (<4 weeks).

This phase I study of pegylated IFN-α-2a (PEG-IFN, Pegasys) in CML patients is similar to a study of the same molecule in patients with renal cell carcinoma (15). In the latter study, the MTD for the first 28 days of treatment was 540 μg/wk. In our study, doses up to 630 μg/wk were not associated with DLT in the first 28 days of therapy. The difference between the two studies may be due to the different patient populations. Our patients with CML had a better performance status, and there were fewer concerns with neutropenia in our study. Assessment of DLT in the first 28 days of treatment may not represent the “true” nature of IFN therapy–based toxicity. Because the treatment is chronic in nature, it should be evaluated after a longer period of follow-up. Based on the experience using PEG-IFN in a phase II study of patients with renal cell carcinoma (18), we designed this study to provide information on the long-term follow-up of patients. Not surprisingly, over the entire 48 weeks of the study period, 96% of the patients receiving PEG-IFN as single-agent therapy required dose reductions because of side effects.

The most significant toxicities reported in this study were similar to those seen with unmodified IFN-α. These included constitutional symptoms of fatigue, fever, myalgias, arthralgias, and hematologic toxicities of anemia and thrombocytopenia (19, 20). Hepatic toxicity with significant liver enzyme elevations occurred in 5 of the 43 patients. Liver enzyme elevation was also the DLT in a patient who took part in the renal cell carcinoma phase I study (15). However, whether PEG-IFN is better tolerated than the standard IFN cannot be addressed in this study and has now been the subject of a randomized study (21).

Peg-Intron (Pegylated IFN α2b, Schering-Plough, Kenilworth, NJ), another type of pegylated IFN, has already been evaluated in patients with CML (22). In a previous study, we showed that Peg-Intron, given at a dose that was a function of body weight, could be tolerated up to a dose of 9 μg/kg/wk, which is very similar to the maximum dose given in the current study. In this study, we also evaluated a combination of Pegasys and low-dose ara-C. This combination was tested because of the frequent utilization of IFN-α and low-dose ara-C in the treatment of CML (23).

In this study, two combination treatment schedules were tested. In one, ara-C was given at a dose of 10 mg/d on a continuous basis. In the other, ara-C was given at 20 mg/m² daily for 10 days of every month. Regardless of the dose of ara-C, an MTD was reached within the first 28 days of therapy with PEG-IFN at a dose of 540 μg/wk. These results were similar to previous findings. Our studies with Peg-Intron in combination with low dose ara-C (10 mg/d) have also shown a need to lower the IFN dose below the MTD to 4.5 to 6 μg/kg/wk (23). The pharmacokinetic characteristics of PEG-IFN were similar to those reported in a phase I study in renal cell carcinoma (15). After s.c. administration of the first dose of PEG-IFN, serum concentration increased rapidly and peaked at 48 hours. Levels were sustained at near-peak levels for an additional 5 days. The pharmacokinetic profile of PEG-IFN contrasts that of the unmodified IFN molecule that has a short serum half-life and rapid clearance. Of interest is the lack of increase in C_max at doses in excess of 450 μg/wk, a finding similar to that seen in the renal cell carcinoma phase I study (15). Although its significance is not defined as of yet, it raises the question of whether dose escalation beyond 450 μg/wk has therapeutic value. Repeated administration of weekly PEG-IFN resulted in a 3- to 5-fold increase in serum levels over the predose levels, and reached steady state after 5 to 9 weeks. Coadministration of ara-C and PEG-IFN did not alter PEG-IFN pharmacokinetics. Serum neopterin levels increased after administration of PEG-IFN and peaked at a PEG-IFN dose of 540 μg/wk. Levels of neopterin were sustained over time with repeated PEG-IFN administration (Fig. 1B) but did not show accumulation as seen up to 5 to 9 weeks with PEG-IFN levels. Changes in WBC are not a very reliable measure for IFN action in patients with CML because of interpatient variation and the presence of hematologic resistance in some CML patients. Nevertheless, a maximum mean percentage drop in WBC increased with dose escalation from 270 to 450 μg/wk. This further supports the notion that a plateau in serum levels and biological activity is reached at a dose of 450 to 540 μg/wk. The interpatient variability in serum concentrations of PEG-IFN and neopterin was high at any dose group and data has been described primarily by descriptive statistics with no attempt for testing any hypothesis. Clinical activity in the form of cytogenetic responses (three complete, one partial, and five minor responses with monotherapy) is of great significance because most patients enrolled in the study had CML with IFN-α resistance, or had shown cytogenetic relapse while on therapy with IFN-α. Furthermore, this patient population had average disease duration of >4 years, and cytogenetic responses have not previously been seen with IFN in these patients. Improved activity with PEG-IFN over standard IFN-α-2a was shown in the hepatitis C studies (13, 14), and Pegasys has been approved in the United States and other countries for that indication. Nevertheless, the relevance of such activity in malignancies is not clear. The suggestion in this study of improved activity for PEG-IFN in patients with CML has been confirmed in a randomized study of Pegasys versus IFN-α-2a in previously untreated patients (21).

Follow up at 1 year after the last patient was entered showed a significantly higher rate of complete hematologic response for PEG-IFN-α-2a (40 kDa) compared with IFN-α-2a (69% versus 41%; P = 0.0008), as well as significantly higher major cytogenetic response (35% versus 18%; P = 0.0162). The 2-year follow-up shows that 10% of patients in the PEG-IFN-α-2a (40 kDa) group had died, compared with 14% of those treated with IFN-α-2a. Another study of Peg-Intron in a newly diagnosed population did not show statistical noninferiority of pegylated IFN to IFN-α-2b (24). This implies that sustained exposure to IFN may overcome resistance to IFN-α in some cases, and that in some patients, resistance to therapy is pharmacologic in nature. One possible explanation for the ability of PEG-IFN to overcome resistance to IFN-α may be the sustained induction of Stat1 (25). This may mediate some of the growth inhibition and proapoptotic activities of IFN-α. The recent introduction of imatinib mesylate as the first line of therapy for CML has altered the treatment algorithms and prognosis of patients with CML. However, at this point, there is no evidence to support a curative potential of imatinib in CML (26). Furthermore, resistance to therapy may develop in patients treated with imatinib (27, 28). Thus, strategies incorporating therapeutic combinations are needed, and PEG-IFN seems to have advantages over standard IFN for such combinations. At the 2003 meeting of the European Hematology Association, Hochhaus et al. (29) reported on a phase I/II dose-finding study of the combination of Pegasys and imatinib, conducted in 32 patients. This study used 300 and 400 mg imatinib and 90 or 180 μg Pegasys in various combinations. Predominant dose-limiting toxicities were grade 3 or 4 cytopenias, reported in 13 of 32 patients, but this was less frequent (one of seven patients) if Pegasys is started after at least 6 weeks of imatinib monotherapy. The report concluded that consecutive treatment, or the combination of 400 mg/d imatinib and 180 μg/wk Pegasys (starting after 6 weeks) warrants further study.