Purpose: We investigated the use of graded-dose peginterferon α-2b (Peg-IFN) in patients with stage IV melanoma overexpressing basic fibroblast growth factor (FGF-2). The primary objective was suppression of plasma FGF-2 to within reference range (≤7.5 pg/mL).

Experimental Design: Plasma FGF-2 was measured at baseline (step 1), and patients with concentrations of 15 pg/mL or more were eligible for study treatment (step 2). Peg-IFN was given weekly at a starting dose of 0.5 μg/kg/wk with increment every 3 weeks based on serial FGF-2 concentrations.

Results: Two hundred seven patients entered step 1; 45 (22%) overexpressed FGF-2 (median = 22 pg/dL). Twenty-nine eligible patients entered step 2 and received treatment. Patients' median age was 64 years (range, 29–84 years). Most had more than two prior therapies. FGF-2 decreased in 28 (97%) patients, with suppression to reference range in 10 (35%). Median time to FGF-2 suppression was 30 days. The best clinical responses were partial response (7%) and stable disease (17%). Median progression-free survival (PFS) and overall survival (OS) were 2.0 and 9.7 months, respectively. Patients who achieved FGF-2 suppression were more likely than those who did not to have a response or stable disease (P = 0.03). VEGF concentrations decreased in 27 patients (93%) during treatment and paralleled those of FGF-2 over time. We found no compensatory increase in VEGF among those with FGF-2 suppression.

Conclusions: Graded-dose Peg-IFN suppresses FGF-2 in patients with metastatic melanoma who overexpress FGF-2. Over one third of patients had complete suppression of plasma FGF-2, which correlated with clinical response to this therapy. Clin Cancer Res; 19(23); 6597–604. ©2013 AACR.

Translational Relevance

To our knowledge, this is the first study that reports the use of peginterferon α-2b (Peg-IFN) as an antiangiogenic agent in melanoma using an individualized graded-dosing approach based on real-time serial measurements of blood plasma angiogenic factors. We were able to demonstrate that in patients overexpressing basic fibroblast growth factor (FGF-2), suppression of FGF-2 was associated with clinical response or stable disease. Future clinical trials may now rationally consider strategies that exploit the antiangiogenic effects of Peg-IFN in combination with molecularly targeted agents and immunotherapy, as well as chemotherapy.

Unlike normal melanocytes, melanoma cells express basic fibroblast growth factor (FGF-2) and its receptor (1). Melanocytes transduced to overexpress FGF-2 transform into a phenotype similar to melanoma (1). Antisense targeting of FGF-2 inhibits angiogenesis and induces melanoma regression in nude mice (2). Other angiogenic factors such as VEGF and interleukin-8 (IL-8) are upregulated during the course of melanoma progression and metastasis (3, 4). In patients with melanoma, increased tumor microvascular density and elevated concentrations of circulating angiogenic factors, including FGF-2, VEGF, and IL-8, are potential markers of worse overall prognosis and abbreviated progression-free survival (PFS; refs. 5, 6). Therefore, targeting of angiogenic factors, including FGF-2, VEGF, and IL-8, seems to be rational approaches to therapy.

IFN-α-2b, currently used in adjuvant therapy and for treatment of metastatic disease, is a cytokine with pleiotropic effects (7). Traditionally, when used for its antiproliferative and immunomodulatory effects, high doses (5–20 million units/m2, 2–3 times weekly) have been favored. Preclinical studies suggest that lower doses of IFN may have antiangiogenic activity through downregulation of the expression of FGF-2, VEGF, IL-8, and matrix metalloproteinases (8, 9). In vitro, low-dose IFN-α inhibits B16 melanoma cell proliferation by interference with an FGF-2 autocrine growth circuit (10). Animal studies show that regulation of angiogenesis by IFN-α is schedule dependent, and that low-dose daily administration induces the most significant antiangiogenic effects and tumor inhibition (8, 11). Furthermore, early clinical studies in vascular tumors, such as infantile hemangioma, giant cell tumor of the mandible, and renal carcinoma, support the selection and monitoring of IFN therapy in relation to elevated plasma or urine FGF concentrations (12–14). Recently, a longer-acting formulation of IFN, peginterferon α-2b (Peg-IFN), was approved by the U.S. Food and Drug Administration for the adjuvant treatment of patients with melanoma at a dose of 6 μg/kg (induction for 8 weeks), then 3 μg/kg (maintenance up to 5 years) injected subcutaneously once a week on the basis of a phase III trial (15, 16).

In this phase II trial, we investigated the use of a low-dose regimen of Peg-IFN as an antiangiogenic agent in patients with metastatic melanoma and elevated plasma concentration of FGF-2. The primary objective of the study was to determine whether circulating FGF-2 concentration can be suppressed to within reference range with Peg-IFN. The secondary objectives were to determine the efficacy and safety profile of Peg-IFN, as well as the association between changes in angiogenic factors (FGF-2 and VEGF) and antitumor activity.

Patient selection

To be eligible, patients must have been at least 18 years of age, had stage IV melanoma from any primary site, and had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2. They might have been previously untreated or could have received up to 3 prior systemic therapies for metastatic disease. Previous treatment with IFN was allowed. Exclusion criteria included active second cancer, history of myocardial infarction within the prior 6 months, history of severe depression, and current pregnancy. Patients with central nervous system (CNS) involvement were allowed, provided that CNS-directed therapy was given and the CNS disease had been clinically stable for at least 3 months. All patients were required to have a baseline measurement of plasma FGF-2 at study for preregistration. Those who overexpressed FGF-2 (≥15 pg/mL) were allowed to go on to register and receive study treatment within 3 working days of registration.

Study design

The treatment schema is shown in Fig. 1. The induction phase corresponded to the period from the beginning of Peg-IFN therapy until the documented suppression of plasma FGF-2 concentration to within reference range (<7.5 pg/mL) was confirmed by two consecutive determinations at least 3 weeks apart. Plasma FGF-2 concentrations were measured and Peg-IFN dose adjusted every 3 weeks. The starting dose of Peg-IFN was 0.5 μg/kg/wk administered subcutaneously. After 3 weeks, the same dose was given if the plasma FGF-2 concentration had declined by 30% or more, or if it had been suppressed to within reference range. Peg-IFN dose was increased every 3 weeks if plasma FGF-2 concentration had decreased by less than 30% and remained above reference range. The six Peg-IFN doses designated in this study were 1 (0.5 μg/kg/wk), 2 (1.0 μg/kg/wk), 3 (2.0 μg/kg/wk), 4 (3.0 μg/kg/wk), 5 (4.0 μg/kg/wk), and 6 (5.0 μg/kg/wk). Induction treatment was continued until plasma FGF-2 concentration was suppressed for two consecutive determinations or until a maximum dose of 5.0 μg/kg/wk was reached. If there was no disease progression, the patient would then enter the maintenance phase, receiving the same dose of Peg-IFN at the end of induction phase to complete another 12 months of therapy. Subsequent Peg-IFN doses were adjusted on the basis of plasma FGF-2 concentrations measured serially every 6 weeks.

Figure 1.

E2602 study schema.

Figure 1.

E2602 study schema.

Close modal

All toxicities were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (version 2.0). Dose modifications were required only for grades 3 and 4 nonanemia/nonfebrile toxicities. When grade 3 or 4 toxicities occurred, Peg-IFN was held until toxicity returned to grade ≤1. Treatment was then restarted at the next lower dose level. If the toxicity occurred during the induction phase, the retreatment automatically placed the patient into the maintenance phase. Protocol treatment was discontinued if the toxicity did not return to grade ≤1 within 3 weeks or any time there was evidence of clinical disease progression. FGF-2 concentrations were not used to determine clinical response or disease progression.

Study evaluations

Clinical examinations and basic laboratory tests were performed before treatment, every 3 weeks during induction, and every 6 weeks during the maintenance phase. Computed tomography (CT) scans of the chest, abdomen, and pelvis were performed every 9 and 12 weeks during induction and maintenance phases, respectively. Responses were assessed using RECIST (Response Evaluation Criteria in Solid Tumors) v.1 guidelines (17).

Correlative studies

Patients were required to submit a paired sample of plasma and urine at the time of study registration and serially every 3 to 6 weeks during treatment. Plasma FGF-2 was measured in real time, whereas plasma VEGF and urine FGF-2 and VEGF were batched and measured later because the latter values were not used to determine Peg-IFN dosing. For both FGF-2 and VEGF assays, we used ELISA kits manufactured by R&D Systems. The 95th percentile value of serum FGF-2 in healthy individuals set by the assay manufacturer is 7.5 pg/mL. In our experience, the mean value of FGF-2 in healthy individuals (n = 10) was 0.9 pg/mL (range, 0–5.2 pg/mL). We, therefore, arbitrarily used a value of ≥15 pg/mL (2 × reference range) as a criterion for FGF-2 overexpression. To account for interassay variability, aliquoted plasma samples used in the previous assay were paired with the more current sample and assayed simultaneously. All angiogenic factor assays were performed centrally (S.M. Callister and D.A. Jobe) at the Gundersen Microbiology Research Laboratory.

Statistical analysis

The primary endpoint of this trial was the suppression of plasma FGF-2 concentration (≤7.5 pg/mL for 2 consecutive determinations at least 3 weeks apart) with Peg-IFN. It was of interest to test the null hypothesis of 10% FGF-2 suppression rate versus the alternative hypothesis of 30% suppression. On the basis of the sample size of 30 eligible patients, our one-sample binomial test had an 84% power to detect this 20% difference in the proportion of respondents who exhibited suppression of FGF-2 concentrations. This was based on a two-sided type I error of 0.05.

The secondary objectives included evaluating overall survival (OS), PFS, and objective tumor responses. For the objective tumor responses, the associations with the FGF-2/VEGF concentrations in the plasma and urine were assessed. Longitudinally collected FGF-2 and VEGF data were summarized descriptively for each case. The lowest concentrations of FGF-2 and VEGF during the induction phase of treatment were correlated with the objective tumor responses. Objective tumor response was dichotomized into those with or without disease progression under protocol therapy [complete response (CR) + partial response (PR) + stable disease vs. progressive disease]. The Wilcoxon rank-sum test was used for these comparisons. The expected tumor response rate was 20%, so we assumed that there would be 6 responders and 24 nonresponders among the 30 eligible cases. With this sample size, the power would be at least 80% if the standardized difference of FGF-2 concentrations between those who had nonprogressive disease and those with progressive disease were at least 1.2. A one-sided type I error of 0.05 was used.

As an exploratory evaluation, we evaluated a 2 × 2 table of objective tumor responses as nonprogressive/progressive and FGF-2/VEGF as high/low. For FGF-2, the cutoff value of 7.5 pg/mL was used to define “high” and “low” groups. For the dichotomized comparisons, proportions of cases with low FGF-2 were compared between those with and without progressive disease. Fisher exact tests were used to compare the proportions. If the difference in the proportions of cases achieving suppression of FGF-2 concentration between the groups was at least 60% (e.g., 80% vs. 20%), then there would be at least 80% power. A one-sided type I error of 0.05 was used. A similar analysis was carried out for the VEGF concentration using the median value as the cutoff value. Given the exploratory nature of these tests, no adjustments were made for multiple comparisons.

The distributions of OS and PFS were estimated using the Kaplan–Meier method. The landmark analysis method was used to compare the survival curves between those who showed plasma FGF-2 suppression and those who did not using log-rank test (18). We chose 63 days after registration as the landmark time point to allow at least three treatment cycles for response (at least 2 treatment cycles were needed to determine FGF-2 suppression). The survival time in the landmark time was defined as time to event (death or progressive disease) or censoring from the landmark time point. Patients who experienced an event of interest (death in OS and progressive disease for PFS) before the landmark time were excluded from the analysis.

Between September 2003 and June 2011, 207 patients registered into step 1 and were screened for plasma FGF-2 concentrations. Of these, 45 (21.7%) patients overexpressed plasma FGF-2, with a median concentration of 22.0 pg/dL (range, 15.0–216.0 pg/dL). Within the latter group, 32 patients enrolled into step 2 and received treatment (Fig. 1). Three patients who received treatment were subsequently excluded from further analysis, 1 for the lack of a baseline CT scan (only a positron emission tomography scan was obtained) and 2 for outdated baseline FGF-2 measurements (more than 4 weeks before step 2 registration). Although FGF response could not be evaluated in these 3 cases, results of the analysis including these patients for all other efficacy endpoints are similar to those excluding them.

Patient characteristics

The clinical features of the 29 eligible patients are summarized in Table 1. The median age was 64 years (range, 29–84 years), and most patients were men (69%) with cutaneous primary (72.4%) who had lactate dehydrogenase (LDH) concentrations within reference range (51.7%). Patients had a median of 2 prior systemic therapies (range, 0–3) and 8 (27.5%) had been treated with IFN previously.

Table 1.

Clinical features of patients on study

Characteristicn (%)
Men 20 (69.0) 
Women 9 (31.0) 
Race 
 White 29 (100.0) 
ECOG performance status 
 0 13 (44.8) 
 1 16 (55.2) 
Primary tumor site 
 Skin 21 (72.4) 
 Mucosa 2 (6.9) 
 Eye 2 (6.9) 
 Unknown 4 (13.8) 
M classification 
 M1a 2 (6.9) 
 M1b 13 (44.8) 
 M1c 14 (48.3) 
LDH 
 Normal 15 (51.7) 
 Elevated 10 (34.5) 
 Unknown 4 (13.8) 
Characteristicn (%)
Men 20 (69.0) 
Women 9 (31.0) 
Race 
 White 29 (100.0) 
ECOG performance status 
 0 13 (44.8) 
 1 16 (55.2) 
Primary tumor site 
 Skin 21 (72.4) 
 Mucosa 2 (6.9) 
 Eye 2 (6.9) 
 Unknown 4 (13.8) 
M classification 
 M1a 2 (6.9) 
 M1b 13 (44.8) 
 M1c 14 (48.3) 
LDH 
 Normal 15 (51.7) 
 Elevated 10 (34.5) 
 Unknown 4 (13.8) 

Treatment

All of the 29 eligible patients received Peg-IFN according to the study protocol. Of these, 9 received maintenance therapy. The median numbers of induction and maintenance cycles were 3 (range, 1–6 cycles) and 4 (range, 1–9 cycles), respectively. The reasons for the termination of treatment were progressive disease (86.2%), treatment toxicities (6.9%), death from disease (3.4%), and symptomatic deterioration without objective disease progression (3.4%).

Efficacy

Plasma FGF-2 concentration was suppressed for two consecutive measurements at least 3 weeks apart in 10 patients—an FGF-2 suppression rate of 34.5% [95% confidence interval (CI), 17.9%–54.3%; P < 0.001 compared with a null hypothesis of 10% response rate]. The median time to FGF-2 suppression was 29.5 days (range, 14–116 days). The median duration of response was 21 days (range, 21–384 days). The objective tumor response rates were as follows: CR (0%), PR (6.9%), stable disease (17.2%), and progression (75.9%). The nonprogression rate was 24.1% (95% CI, 10.3%–43.5%). The median PFS was 2 months (range, 0.6–13.7 months; 95% CI, 1.8–2.3), and the median OS was 9.7 months (range, 1.7–45.4 months; 95% CI, 4.4–12.5).

The landmark analysis of OS was performed 63 days after registration to compare the OS between those who achieved FGF-2 suppression and those who did not. Three patients were excluded because they died before the landmark time. The median survivals for the suppressed and nonsuppressed groups were 7.8 months (95% CI, 0.1–10.4) and 7.7 months (95% CI, 3.1–14.4), respectively (Fig. 2; P = 0.31). Patients who achieved FGF-2 suppression were more likely than those who did not to have a response or stable disease (50% vs. 10.5%; P = 0.03; Fig. 3A).

Figure 2.

Comparison of OS among those who did and did not achieve suppression of plasma FGF-2.

Figure 2.

Comparison of OS among those who did and did not achieve suppression of plasma FGF-2.

Close modal
Figure 3.

Plasma angiogenic factor concentrations during Peg-IFN therapy. A, best percentage change of angiogenic factor concentrations from baseline relative to tumor response. The orange doughnut-shaped symbol highlights the patients who had suppression of plasma FGF-2. Patients without a visible change in their angiogenic factor had 0% change, consistently undetectable plasma concentrations, or had progressive disease soon after treatment and declined further angiogenic factor measurement. B and C, serial angiogenic factor concentrations over time. D, baseline angiogenic factor concentrations for each patient relative to tumor response. The green, yellow, and orange diamonds represent patients who had partial response, stable disease, and progressive disease, respectively.

Figure 3.

Plasma angiogenic factor concentrations during Peg-IFN therapy. A, best percentage change of angiogenic factor concentrations from baseline relative to tumor response. The orange doughnut-shaped symbol highlights the patients who had suppression of plasma FGF-2. Patients without a visible change in their angiogenic factor had 0% change, consistently undetectable plasma concentrations, or had progressive disease soon after treatment and declined further angiogenic factor measurement. B and C, serial angiogenic factor concentrations over time. D, baseline angiogenic factor concentrations for each patient relative to tumor response. The green, yellow, and orange diamonds represent patients who had partial response, stable disease, and progressive disease, respectively.

Close modal

Adverse events

All 32 patients were included in the adverse event analysis, the results of which are summarized in Table 2. Thirteen (40.6%) patients experienced at least one major (grade ≥3) adverse event, with the most common being fatigue (21.9%), followed by nausea (9.4%), neutropenia (6.3%), and infection without neutropenia (6.3%). One patient died while on therapy associated with progressive disease.

Table 2.

Adverse events in patients

ToxicityAll grades, n (%)Grade ≥3, n (%)
Cardiovascular 
 Dizziness 6 (18.8) 1 (3.1) 
 Hypotension 1 (3.1) 1 (3.1) 
Constitutional 
 Chills 7 (21.9) 1 (3.1) 
 Fatigue 26 (81.3) 7 (21.9) 
Gastrointestinal 
 Constipation 4 (12.5) 1 (3.1) 
 Laboratory—ALT 10 (31.3) 1 (3.1) 
 Nausea 14 (43.8) 3 (9.4) 
 Vomiting 6 (18.8) 1 (3.1) 
 Other 2 (6.3) 1 (3.1) 
Hematologic 
 Anemia 15 (46.9) 1 (3.1) 
 Neutropenia 12 (37.5) 1 (3.1) 
Infectious disease 
 Infection without neutropenia 3 (9.4) 2 (6.3) 
Musculoskeletal 
 Myalgia 9 (28.1) 1 (3.1) 
Neurologic 
 Confusion 2 (6.3) 1 (3.1) 
 Depression 6 (18.8) 1 (3.1) 
 Tremor 2 (6.3) 1 (3.1) 
 Pain 2 (6.3) 1 (3.1) 
All other 29 (90.6) 2 (6.3) 
Total 32 (100) 13 (40.6) 
ToxicityAll grades, n (%)Grade ≥3, n (%)
Cardiovascular 
 Dizziness 6 (18.8) 1 (3.1) 
 Hypotension 1 (3.1) 1 (3.1) 
Constitutional 
 Chills 7 (21.9) 1 (3.1) 
 Fatigue 26 (81.3) 7 (21.9) 
Gastrointestinal 
 Constipation 4 (12.5) 1 (3.1) 
 Laboratory—ALT 10 (31.3) 1 (3.1) 
 Nausea 14 (43.8) 3 (9.4) 
 Vomiting 6 (18.8) 1 (3.1) 
 Other 2 (6.3) 1 (3.1) 
Hematologic 
 Anemia 15 (46.9) 1 (3.1) 
 Neutropenia 12 (37.5) 1 (3.1) 
Infectious disease 
 Infection without neutropenia 3 (9.4) 2 (6.3) 
Musculoskeletal 
 Myalgia 9 (28.1) 1 (3.1) 
Neurologic 
 Confusion 2 (6.3) 1 (3.1) 
 Depression 6 (18.8) 1 (3.1) 
 Tremor 2 (6.3) 1 (3.1) 
 Pain 2 (6.3) 1 (3.1) 
All other 29 (90.6) 2 (6.3) 
Total 32 (100) 13 (40.6) 

Correlative studies

Plasma FGF-2 concentration was suppressed to within reference range in only 10 patients (34.5%), but it decreased in nearly all patients (28 of 29; 97%) during Peg-IFN therapy (Fig. 3A). Plasma VEGF concentrations decreased concurrently in 27 patients (93%; Fig. 3A). No compensatory increase in plasma VEGF concentration was observed among those who had plasma FGF-2 suppressed. Serial values of plasma FGF-2 and VEGF concentrations over time are shown in Fig. 3B and C, respectively. The pretreatment values of plasma FGF-2 and VEGF for patients relative to their tumor response are depicted in Fig. 3D. As a group, the mean posttreatment values were lower than the pretreatment values for both FGF-2 (6.7 vs. 29.5 pg/mL; P = 0.005) and VEGF (37.9 vs. 127.5 pg/mL; P = 0.059). Patients who achieved an objective response or stable disease had similar FGF-2 at baseline (mean, 21.1 vs. 32.2 pg/mL; P = 0.49) but perhaps higher VEGF (261.6 vs. 84.9 pg/mL; P = 0.08) concentrations compared with those who did not. Additional data on angiogenic factor concentrations in the plasma and urine are shown in the Supplementary Table S1.

This phase II ECOG–American College of Radiology Imaging Network (ACRIN) trial studied the antiangiogenic and antitumor effects of Peg-IFN in patients with metastatic melanoma who were selected according to overexpression of FGF-2 in their plasma. We report several pertinent findings of this study. Approximately one fifth of the patients with metastatic melanoma overexpressed FGF-2. Using a graded-dosing Peg-IFN regimen based on real-time serial FGF-2 plasma concentrations, we have documented clinical antitumor activity and the suppression not only of FGF-2 but also of VEGF. Suppression of plasma FGF-2 was found to be associated with the clinical outcome of antitumor response or stable disease. This study did not document longer survival in association with the modulation of FGF-2 in the patients enrolled in this trial, but the number of patients studied did not reach the threshold recommended by Korn and colleagues (19). An increase in plasma VEGF was not observed as a compensatory mechanism when FGF-2 was suppressed by Peg-IFN.

To our knowledge, this is the first study that reports the use of Peg-IFN as an antiangiogenic agent in a graded-dosing approach based on real-time serial measurements of blood plasma angiogenic factors. Previous phase II trials in melanoma have evaluated varied but fixed dosages of Peg-IFN ranging from approximately 0.5 to 6.0 μg/kg/wk as a single agent or in combination with other drugs (dacarbazine, IL-2, sorafenib, thalidomide, and temozolamide). Response rates ranged from 0% to 29%, with higher rates of responses in trials in which chemotherapy agents were included (20–27). Only 1 of these studies measured pre- and posttreatment serum angiogenic factor concentrations, without finding any association with antitumor responses or altered survival outcome. In addition, FGF-2 and VEGF concentrations were not found to change significantly from baseline (26). Similar findings of the dissociation between presence of clinical antitumor response and lack of angiogenic response were noted in several phase II trials using daily or intermittent regimens of recombinant nonpegylated IFN in conjunction with bevacizumab and endostatin (28–31).

In contrast, our study has shown that in the majority (>90%) of patients with metastatic melanoma treated, plasma FGF-2 and VEGF concentrations were reduced by Peg-IFN. Antitumor response and stable disease were more likely among those whose FGF-2 concentrations were suppressed to within reference range than among those whose were not. Several factors, including patient selection and Peg-IFN dosing regimen, may potentially explain the discrepancies between these findings. Unlike previous studies, only patients who overexpressed FGF-2, the presumptive antiangiogenic target of Peg-IFN, were included in our study. Because fewer than a quarter of patients overexpress FGF-2, an antiangiogenic signal may be easily overlooked if the study population is not enriched with the putative biologic target. Our Peg-IFN dosing regimen was novel in that we adjusted the dose based on real-time measurement of plasma FGF-2 concentration to maximize biologic response and minimize toxicity. Preclinical studies suggest that the regulation of angiogenesis by IFN is dose and schedule dependent, with low dose and constant tumor exposure being the most effective (8, 11). Moreover, the optimal biologic dose of Peg-IFN may vary from patient to patient. Therefore, an individualized approach may be more effective.

Multiple lines of evidence support the existence of cross-talk between FGF and VEGF pathways in melanoma progression. Preclinical and clinical studies suggest that upregulation of the FGF pathway may serve as a mechanism of resistance to anti-VEGF therapy and potentially vice versa (32). It is notable that among the patients with complete suppression of FGF-2, we did not observe any compensatory increase in VEGF. This suggests that upregulation of the VEGF pathway may not be a primary escape mechanism when the FGF-2 pathway is targeted in metastatic melanoma. Alternatively, this could be due to concomitant Peg-IFN suppression of VEGF production.

Early-phase clinical trials targeting the FGF ligands and their receptors are in development in melanoma and other cancers (32). A recent phase I/II trial of dovitinib, a small-molecule inhibitor of FGF receptors as well as other class III receptor tyrosine kinases, showed an acceptable safety profile but limited clinical activity (26% stable disease; ref. 33). In our study, the antitumor response rate was low (7% partial response and 7% stable disease) but comparable with the results of another trial using fixed-dose Peg-IFN as a single agent (20). This suggests that FGF or FGF receptor targeting alone may not be sufficient for therapy of metastatic advanced melanoma and that combinations with other targeted agents or chemotherapy may be required to achieve more meaningful results. In support of this, a synergistic drug interaction was observed when combining FGF receptor inhibitors (SU5402, PD166866) with the multikinase/BRAF inhibitor sorafenib or the V600E mutant-specific BRAF inhibitor vemurafenib in a preclinical model targeting melanoma overexpressing FGF-2, FGF-5, and FGF-18. In contrast, the addition of the same FGF receptor inhibitors to dacarbazine showed only modestly increased activity (34).

In conclusion, we have shown that graded dosages of Peg-IFN titrated against plasma FGF-2 exhibit antitumor activity that correlates with suppression of FGF-2 concentrations in patients with refractory metastatic melanoma selected for elevated baseline FGF-2 concentrations. Titration of Peg-IFN based on an individual patient's serial plasma FGF-2 concentrations is feasible and provides a route to optimize dosing of Peg-IFN in melanoma. Future clinical trials may now rationally consider strategies that exploit the antiangiogenic effects of Peg-IFN in combination with molecularly targeted agents and immunotherapy, as well as chemotherapy.

A.A. Tarhini is consultant/advisory board member and received commercial research grant from Merck. J.M. Kirkwood received commercial research grant from Promethus and is consultant/advisory board member of GlaxoSmithKline, Merck, Vical, Bristol–Myers Squibb, and Celgene. No potential conflicts of interest were disclosed by the other authors.

The contents of this study are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.

Conception and design: R.S. Go, S.J. Lee, J.M. Kirkwood

Development of methodology: R.S. Go, J.M. Kirkwood

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R.S. Go, S.M. Callister, D.A. Jobe, R.M. Conry, A.A. Tarhini, J.M. Kirkwood

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R.S. Go, S.J. Lee, D. Shin, S.M. Callister, D.A. Jobe, J.M. Kirkwood

Writing, review, and/or revision of the manuscript: R.S. Go, S.J. Lee, A.A. Tarhini, J.M. Kirkwood

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R.S. Go

Study supervision: R.S. Go, J.M. Kirkwood

This study was conducted by the Eastern Cooperative Oncology Group (R.L. Comis) and supported in part by the Public Health Service Grants CA23318, CA66636, CA21115, and CA39229 and from the National Cancer Institute, NIH, and the Department of Health and Human Services. Additional funding was provided by Gundersen Medical Foundation and Gundersen Health System.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Nesbit
M
,
Nesbit
HK
,
Bennett
J
,
Andl
T
,
Hsu
MY
,
Dejesus
E
, et al
Basic fibroblast growth factor induces a transformed phenotype in normal human melanocytes
.
Oncogene
1999
;
18
:
6469
76
.
2.
Wang
Y
,
Becker
D
. 
Antisense targeting of basic fibroblast growth factor and fibroblast growth factor receptor-1 in human melanomas blocks intratumoral angiogenesis and tumor growth
.
Nat Med
1997
;
3
:
887
93
.
3.
Salven
P
,
Heikkila
P
,
Joensuu
H
. 
Enhanced expression of vascular endothelial growth factor in metastatic melanoma
.
Br J Cancer
1997
;
76
:
930
4
.
4.
Singh
RK
,
Varney
ML
. 
IL-8 expression in malignant melanoma: implications in growth and metastasis
.
Histol Histopathol
2000
;
15
:
843
9
.
5.
Vlaykova
T
,
Laurila
P
,
Muhonen
T
,
Hahka-Kemppinen
M
,
Jekunen
A
,
Alitalo
K
, et al
Prognostic value of tumour vascularity in metastatic melanoma and association of blood vessel density with vascular endothelial growth factor expression
.
Melanoma Res
1999
;
9
:
59
68
.
6.
Ugurel
S
,
Rappl
G
,
Tilgen
W
,
Reinhold
U
. 
Increased serum concentration of angiogenic factors in malignant melanoma patients correlates with tumor progression and survival
.
J Clin Oncol
2001
;
19
:
577
83
.
7.
Jonasch
E
,
Haluska
FG
. 
Interferon in oncological practice: review of interferon biology, clinical applications, and toxicities
.
Oncologist
2001
;
6
:
34
55
.
8.
Slaton
JW
,
Perrotte
P
,
Inoue
K
,
Dinney
CP
,
Fidler
IJ
. 
Interferon-alpha-mediated down-regulation of angiogenesis-related genes and therapy of bladder cancer are dependent on optimization of biological dose and schedule
.
Clin Cancer Res
1999
;
5
:
2726
34
.
9.
Salven
P
,
Anttonen
K
,
Repo
H
,
Joensuu
H
,
Orpana
A
. 
Endotoxins induce and interferon alpha suppresses vascular endothelial growth factor (VEGF) production in human peripheral blood mononuclear cells
.
FASEB J
2001
;
15
:
1318
20
.
10.
Torcia
M
,
Lucibello
M
,
De Chiara
G
,
Labardi
D
,
Nencioni
L
,
Bonini
P
, et al
Interferon-alpha-induced inhibition of B16 melanoma cell proliferation: interference with the bFGF autocrine growth circuit
.
Biochem Biophys Res Commun
1999
;
262
:
838
44
.
11.
Ozawa
S
,
Shinohara
H
,
Kanayama
HO
,
Bruns
CJ
,
Bucana
CD
,
Ellis
LM
, et al
Suppression of angiogenesis and therapy of human colon cancer liver metastasis by systemic administration of interferon-alpha
.
Neoplasia
2001
;
3
:
154
64
.
12.
Chang
E
,
Boyd
A
,
Nelson
CC
,
Crowley
D
,
Law
T
,
Keough
KM
, et al
Successful treatment of infantile hemangiomas with interferon-alpha-2b
.
J Pediatr Hematol Oncol
1997
;
19
:
237
44
.
13.
Vermeulen
PB
,
Dirix
LY
,
Martin
M
,
Lemmens
J
,
Van Oosterom
AT
. 
Serum basic fibroblast growth factor and vascular endothelial growth factor in metastatic renal cell carcinoma treated with interferon alfa-2b
.
J Natl Cancer Inst
1997
;
89
:
1316
7
.
14.
Kaban
LB
,
Mulliken
JB
,
Ezekowitz
RA
,
Ebb
D
,
Smith
PS
,
Folkman
J
. 
Antiangiogenic therapy of a recurrent giant cell tumor of the mandible with interferon alfa-2a
.
Pediatrics
1999
;
103
:
1145
9
.
15.
Eggermont
AM
,
Suciu
S
,
Santinami
M
,
Testori
A
,
Kruit
WH
,
Marsden
J
, et al
Adjuvant therapy with pegylated interferon alfa-2b versus observation alone in resected stage III melanoma: final results of EORTC 18991, a randomised phase III trial
.
Lancet
2008
;
372
:
117
26
.
16.
Herndon
TM
,
Demko
SG
,
Jiang
X
,
He
K
,
Gootenberg
JE
,
Cohen
MH
, et al
U.S. Food and Drug Administration approval: peginterferon-alfa-2b for the adjuvant treatment of patients with melanoma
.
Oncologist
2012
;
17
:
1323
8
.
17.
Therasse
P
,
Arbuck
SG
,
Eisenhauer
EA
,
Wanders
J
,
Kaplan
RS
,
Rubinstein
L
, et al
New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada
.
J Natl Cancer Inst
2000
;
92
:
205
16
.
18.
Anderson
JR
,
Cain
KC
,
Gelber
RD
. 
Analysis of survival by tumor response
.
J Clin Oncol
1983
;
1
:
710
9
.
19.
Korn
EL
,
Liu
PY
,
Lee
SJ
,
Chapman
JA
,
Niedzwiecki
D
,
Suman
VJ
, et al
Meta-analysis of phase II cooperative group trials in metastatic stage IV melanoma to determine progression-free and overall survival benchmarks for future phase II trials
.
J Clin Oncol
2008
;
26
:
527
34
.
20.
Dummer
R
,
Garbe
C
,
Thompson
JA
,
Eggermont
AM
,
Yoo
K
,
Maier
T
, et al
Randomized dose-escalation study evaluating peginterferon alfa-2a in patients with metastatic malignant melanoma
.
J Clin Oncol
2006
;
24
:
1188
94
.
21.
Hwu
WJ
,
Panageas
KS
,
Menell
JH
,
Lamb
LA
,
Aird
S
,
Krown
SE
, et al
Phase II study of temozolomide plus pegylated interferon-alpha-2b for metastatic melanoma
.
Cancer
2006
;
106
:
2445
51
.
22.
Vaishampayan
UN
,
Heilbrun
LK
,
Marsack
C
,
Smith
DW
,
Flaherty
LE
. 
Phase II trial of pegylated interferon and thalidomide in malignant metastatic melanoma
.
Anticancer Drugs
2007
;
18
:
1221
6
.
23.
Hauschild
A
,
Dummer
R
,
Ugurel
S
,
Kaehler
KC
,
Egberts
F
,
Fink
W
, et al
Combined treatment with pegylated interferon-alpha-2a and dacarbazine in patients with advanced metastatic melanoma: a phase 2 study
.
Cancer
2008
;
113
:
1404
11
.
24.
Spieth
K
,
Kaufmann
R
,
Dummer
R
,
Garbe
C
,
Becker
JC
,
Hauschild
A
, et al
Temozolomide plus pegylated interferon alfa-2b as first-line treatment for stage IV melanoma: a multicenter phase II trial of the Dermatologic Cooperative Oncology Group (DeCOG)
.
Ann Oncol
2008
;
19
:
801
6
.
25.
Recchia
F
,
Candeloro
G
,
Necozione
S
,
Fumagalli
L
,
Bratta
M
,
Rea
S
. 
Multicenter phase II study of chemoimmunotherapy in the treatment of metastatic melanoma
.
Anticancer Drugs
2008
;
19
:
201
7
.
26.
Guillot
B
,
Khamari
A
,
Cupissol
D
,
Delaunay
M
,
Bedane
C
,
Dreno
B
, et al
Temozolomide associated with PEG-interferon in patients with metastatic melanoma: a multicenter prospective phase I/II study
.
Melanoma Res
2008
;
18
:
141
6
.
27.
Egberts
F
,
Gutzmer
R
,
Ugurel
S
,
Becker
JC
,
Trefzer
U
,
Degen
A
, et al
Sorafenib and pegylated interferon-alpha2b in advanced metastatic melanoma: a multicenter phase II DeCOG trial
.
Ann Oncol
2011
;
22
:
1667
74
.
28.
Varker
KA
,
Biber
JE
,
Kefauver
C
,
Jensen
R
,
Lehman
A
,
Young
D
, et al
A randomized phase 2 trial of bevacizumab with or without daily low-dose interferon alfa-2b in metastatic malignant melanoma
.
Ann Surg Oncol
2007
;
14
:
2367
76
.
29.
Vihinen
PP
,
Hernberg
M
,
Vuoristo
MS
,
Tyynela
K
,
Laukka
M
,
Lundin
J
, et al
A phase II trial of bevacizumab with dacarbazine and daily low-dose interferon-alpha2a as first line treatment in metastatic melanoma
.
Melanoma Res
2010
;
20
:
318
25
.
30.
Grignol
VP
,
Olencki
T
,
Relekar
K
,
Taylor
C
,
Kibler
A
,
Kefauver
C
, et al
A phase 2 trial of bevacizumab and high-dose interferon alpha 2B in metastatic melanoma
.
J Immunother
2011
;
34
:
509
15
.
31.
Moschos
SJ
,
Odoux
C
,
Land
SR
,
Agarwala
S
,
Friedland
D
,
Volker
KM
, et al
Endostatin plus interferon-alpha2b therapy for metastatic melanoma: a novel combination of antiangiogenic and immunomodulatory agents
.
Melanoma Res
2007
;
17
:
193
200
.
32.
Lieu
C
,
Heymach
J
,
Overman
M
,
Tran
H
,
Kopetz
S
. 
Beyond VEGF: inhibition of the fibroblast growth factor pathway and antiangiogenesis
.
Clin Cancer Res
2011
;
17
:
6130
9
.
33.
Kim
KB
,
Chesney
J
,
Robinson
D
,
Gardner
H
,
Shi
MM
,
Kirkwood
JM
. 
Phase I/II and pharmacodynamic study of dovitinib (TKI258), an inhibitor of fibroblast growth factor receptors and VEGF receptors, in patients with advanced melanoma
.
Clin Cancer Res
2011
;
17
:
7451
61
.
34.
Metzner
T
,
Bedeir
A
,
Held
G
,
Peter-Vorosmarty
B
,
Ghassemi
S
,
Heinzle
C
, et al
Fibroblast growth factor receptors as therapeutic targets in human melanoma: synergism with BRAF inhibition
.
J Invest Dermatol
2011
;
131
:
2087
95
.