Adjuvant Interferon Therapy for Patients with Uveal Melanoma at High Risk of Metastasis

Adjuvant Interferon Therapy for Patients with Uveal Melanoma at High Risk of Metastasis Anne Marie Lane, MPH,1 Kathleen M. Egan, ScD,2 David Harmon, MD,3 Amy Holbrook, MS,1 John E. Munzenrider, MD,4 Evangelos S. Gragoudas, MD1 Purpose: To examine whether interferon (IFN)-alfa-2a treatment after radiation or enucleation reduces death rates in patients with uveal melanoma. Design: Interventional, comparative case series. Participants: Subjects were identified through the ocular oncology clinic of the Massachusetts Eye and Ear Infirmary. Patients eligible for the study were at increased risk of metastasis because of the presence of at least one of the following characteristics: age ⱖ65 years, largest tumor diameter (LTD) ⱖ15 mm, ciliary body involvement of the tumor, or extrascleral tumor extension. Methods: Between May 1995 and June 1999, 121 patients with choroidal or ciliary body melanoma began a 2-year course of therapy (3 MIU IFN-alfa-2a subcutaneously 3 times per week), initiated within 3 years of primary therapy. All patients underwent regular monitoring for drug toxicity. To evaluate IFN-alfa-2a efficacy, we selected a series of historical controls frequency-matched (2:1) to IFN-alfa-2a–treated patients on age (⫾5 years), LTD (⫾3 mm), gender, and survival time between primary therapy and initiation of IFN therapy. Survival status was ascertained for all patients through December 2006. Main Outcome Measures: Melanoma-related mortality, metastasis, IFN-related toxicities. Results: Fifty-five patients (45%) completed therapy; the median dose for IFN-alfa-2a–treated patients was 792 MIU (85% of the theoretic dose). The median follow-up time in the IFN-alfa-2a–treated group was approximately 9 years. Treatment and control groups were similar with respect to age (P ⫽ 0.78), LTD (P ⫽ 0.38), and gender (P ⫽ 1.0). Of 363 patients, 108 developed metastasis under observation; 42 of these were IFN-alfa-2a– treated patients. Cumulative 5-year melanoma-related death rates were 17% in the radiation or enucleation-only group, 15% in those who completed the entire IFN-alfa-2a course, and 35% in those who discontinued IFN-alfa-2a therapy. In multivariate Cox regression, IFN-alfa-2a had no significant influence on melanoma-related mortality (rate ratio ⫽ 1.02, 95% confidence interval, 0.68 –1.5, P ⫽ 0.91) or all-cause mortality (rate ratio ⫽ 0.84, 95% confidence interval, 0.58 –1.2, P ⫽ 0.34). Conclusions: Interferon-alfa-2a has no material influence on survival in patients with choroidal melanoma. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references. Ophthalmology 2009;116:2206 –2212 © 2009 by the American Academy of Ophthalmology.

Uveal melanoma is the most common primary cancer of the eye and often results not only in vision loss but also in metastatic death. Regardless of the primary treatment modality, radiotherapy or enucleation of the eye, risk of metastasis is increased and prognosis is poor in patients of more advanced age, with large tumors, or with tumors that involve the ciliary body or exhibit extrascleral extension. Currently there is no effective treatment for metastatic uveal melanoma. Metastasis occurs in up to 50%1 of patients with high-risk features after proton irradiation despite rates of local control approaching 100%, suggesting that many patients have clinically undetectable micrometastasis at the time of diagnosis. Class I interferons (IFNs), which include IFN-alfa, IFNbeta, and IFN-gamma, are a class of cytokines produced by leukocytes in response to viruses and other foreign substances in the body. In addition to their antiviral properties, these IFNs alter the immune response and inhibit cell pro-

2206

© 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.

liferation in malignant and nonmalignant cells. Recombinant IFN-alfa, with 2 subtypes, 2a and 2b, is engineered using DNA from human leukocytes that codes for the protein and has been approved by the United States Food and Drug Administration to treat chronic hepatitis C and chronic hepatitis B (2b only), as well as several cancers including hairy cell leukemia, acquired immune deficiency syndromerelated Kaposi’s sarcoma, chronic phase Philadelphia chromosome positive chronic myelogenous leukemia (2a only), and primary malignant melanoma (2b only). In clinical trials, IFN has been evaluated alone or in combination with other chemotherapeutics in the treatment of metastatic pancreatic carcinoma,2 renal cell carcinoma,3– 6 cutaneous melanoma,7–9 and uveal melanoma.10 –14 The antiproliferative and immunomodulatory effects of IFN-alfa in primary uveal melanoma cell lines have been demonstrated in vitro.15 However, clinical trials have generally not supported efficacy of IFN-alfa in the treatment of ISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2009.04.044

Lane et al 䡠 Adjuvant Interferon Therapy for Uveal Melanoma Table 1. Inclusion Criteria: Systemic Measures of Health Hematology Profile

Chemistry Profile

Absence of

Hemoglobin ⬎11 g/dl WBC ⬎4000/ml Platelet count ⬎150,000/ml

Serum creatinine ⬍1.7 mg/dl Total bilirubin ⬍1.3 mg/dl Liver enzymes (AST/ALT) ⬍1.5 times upper limit of normal

Uncontrolled hypertension Autoimmune disease Clinical depression Hepatic, renal, CNS disease Cardiovascular disease NYHA functional capacity of III-IV

TSH ⬎0.4, ⬍9.0 uIU/ml

ALT ⫽ alanine aminotransferase; AST ⫽ aspartate aminotransferase; CNS ⫽ central nervous system; NYHA ⫽ New York Heart Association; TSH ⫽ thyroid stimulating hormone; WBC ⫽ white blood cell.

metastatic uveal melanoma. Pyrhonen and colleagues reported a 15% partial response rate in a small (N ⫽ 22) phase II study of bleomycin, vincristine, lomustine, and dacarbazine, and human leukocyte IFN-alpha for patients with stage IVB uveal melanoma,13 but another similar study11 found this combination therapy to be ineffective (no patients exhibited a complete or partial response). An approximate 14% response rate (partial and complete) was reported in patients treated with fotemustine, IFN-alpha, and interleukin-2.10 In another study, disease stabilization was achieved in 1 of 6 patients with metastatic uveal melanoma treated with a combination of thalidomide and IFN-alpha-2b who had failed other treatments.14 This lack of efficacy may be due to inopportune timing of the intervention. In patients with clinically established metastases, tumor cells have eluded the body’s immune response and established a vascular network to grow, and therefore are likely to be capable of resisting the effects of immunotherapy. Earlier intervention may be necessary to inhibit the proliferation and migration of malignant cells. The use of IFN as an adjuvant to primary therapy has been limited in the clinical setting to the treatment of patients with high-risk resected cutaneous melanoma.16,17 To our knowledge, no studies have examined the efficacy of adjunctive IFN therapy in prolonging patient survival in uveal melanoma. We addressed this question in a series of patients undergoing proton therapy or enucleation of the eye.

Materials and Methods This was a nonrandomized trial in which all enrolled patients received IFN-alfa-2a, recombinant (Roferon-A, Roche Laboratories, Nutley, NJ). The study was conducted according to the tenets of the Declaration of Helsinki and approved by the Human Subjects Committee of the Massachusetts Eye and Ear Infirmary. All patients provided written informed consent to participate in the study. Between May 1995 and June 1999, all patients presenting to the Massachusetts Eye and Ear Infirmary for primary treatment or follow-up examination of a choroidal or ciliary body melanoma who met study eligibility criteria were invited to undergo a 2-year course of IFN therapy. To be eligible, patients had 1 or more of the following characteristics: age ⱖ65 years, largest tumor diameter (LTD) ⱖ15 mm, ciliary body involvement of the tumor, extrascleral tumor extension, and diagnosis within 3 years of the study enrollment date. The latter period corresponds to the time of

greatest risk of metastasis for these patients. Because of the known toxicities associated with IFN, patients who entered the study also had to be in good general health, with acceptable hematology, chemistry, and medical history profiles (Table 1). Patients with evidence of metastatic disease or other primary malignancy based on physical examination or liver function studies, or who had previously used chemotherapeutics, systemic steroids, immunomodulators, or antineoplastic drugs, were excluded from the study. Of 347 eligible patients, 215 (61%) declined and 121 (35%) enrolled in the study. Of these, 118 patients were treated with proton irradiation and 3 patients underwent enucleation as primary treatment for the tumor. Adjuvant IFN therapy was initiated within 3 years of irradiation or enucleation, the majority of patients (80%) commencing IFN treatment immediately after diagnosis and primary treatment for the tumor. The IFN treatment protocol consisted of 3 MIU IFN administered 3 times per week (TIW) by subcutaneous injection over a 2-year period. Interferon treatment was self-administered at home after training and first injection in the retina service at Massachusetts Eye and Ear Infirmary by a study nurse, who also followed the patient for compliance and adverse effects. All patients underwent regular monitoring for drug toxicity. To evaluate IFN efficacy, for each IFN-treated patient, we selected 2 patients treated at the Massachusetts Eye and Ear Infirmary before initiation of the study of the same gender and of similar age (⫾5 years) and LTD (⫾3 mm). To avoid survivor bias, selected patients were also matched on the survival time between primary therapy and initiation of IFN. These historical controls were similar to treated patients with regard to pretreatment workup, diagnostic tests performed, and number of follow-up examinations completed. All patients were treated by a single physician (EG). Survival status was ascertained for all patients using the Social Security Death Index or by active surveillance through December 2006. Cause of death was confirmed by hospital records, death certificates, or reports by the patient’s next-of-kin. Median follow-up time was approximately 9.5 years among surviving patients in the IFN group. To avoid bias, we censored follow-up time in the surviving controls to 9.5 years. The primary end point was diagnosis of melanoma-related mortality. Safety end points included laboratory and clinical assessments for abnormalities, such as thrombocytopenia, neutropenia, elevated transaminases, hypothyroidism, depression, and diabetes. The Kaplan–Meier method was used to determine rates of melanoma-related and all-cause mortality, and survival distributions were compared according to treatment group by the log-rank test. Cox regression analysis was completed to evaluate the influence of IFN treatment on patient outcome while controlling for other prognostic factors. Descriptive statistics, including fre-

2207

Ophthalmology Volume 116, Number 11, November 2009 quency, duration, and outcome of all safety end points, were calculated among patients in the IFN-treated group.

Results Interferon Treatment Of the 121 patients who enrolled in the study and began a course of adjuvant therapy, 55 (45%) completed therapy and 66 (55%) discontinued therapy, with a median administered dose of 792 MIU (84.6% of theoretic dose). Adverse events, such as symptoms or onset of new illnesses (n ⫽ 25, 37.9%) and abnormal laboratory values (n ⫽ 23, 34.8%), were the most common reasons for discontinuing therapy. Exacerbation of a preexisting illness and development of metastasis each accounted for the premature withdrawal of 8 patients (12.1%).

Figure 1. Unadjusted rates of melanoma-related mortality by treatment group. IFN ⫽ interferon; PBI ⫽ proton beam irradiation.

Interferon Toxicities Dose-limiting toxicities occurred in 28 patients and included thrombocytopenia (platelet count 75,000 –100,000/m) in 3 cases (10.7%), elevated liver enzymes (aspartate aminotransferase or alanine aminotransferase ⬎ 2.5 times the upper limit of normal) in 12 cases (42.9%), and thyroid function alterations in 10 cases (35.7%). In 10 cases, therapy was interrupted 1 to 3 weeks until laboratory values returned to normal, at which point therapy was reintroduced. Two patients were able to complete therapy after such an interruption, 5 patients had to discontinue therapy when toxicity recurred after IFN was reintroduced, and 3 patients discontinued therapy because of other adverse effects. Eighteen patients discontinued therapy when hematology or chemistry values initially reached toxic values. Patients and physicians reported 1659 adverse effects, with 129 reports of fatigue/lethargy (7.8%), 79 incidents of headache (4.8%), and 69 reports of chills (4.2%). Arthralgias and depression each accounted for approximately 3% of all adverse events. All patients reported flu-like symptoms after the first few injections. Most adverse events were mild to moderate in severity and resolved with continued IFN therapy.

Mortality Patients treated with adjuvant therapy after radiation or enucleation were generally similar to patients treated with radiation or enucleation only with regard to baseline demographic characteristics and factors predictive of tumor-related mortality (Table 2). The 2 treatment groups were well balanced with regard to age, gender, LTD, and prevalence of extrascleral extension. Patients Table 2. Demographic and Prognostic Factors by Treatment Group PBI or PBI or Enucleation ⴙ IFN Enucleation P Value Age (median) LTD (median) Male (%) CB involvement (%) Extrascleral extension (%)

63 yrs 15 55 55 8

64 yrs 14 55 38 6

.78 .38 1.0 .002 .38

CB ⫽ ciliary body; IFN ⫽ interferon-alfa-2a, recombinant; LTD ⫽ largest tumor diameter; PBI ⫽ proton beam irradiation.

2208

who received adjuvant therapy were more likely to have a tumor that involved the ciliary body. Among IFN-treated patients, 42 (34.7%) developed metastasis compared with 66 patients (27.5%) in the control group. Seventeen patients (31%) developed metastasis after completing the 2-year course of IFN therapy, and 13 of these died of the disease while under observation. Of the 66 patients who discontinued adjuvant therapy, 25 (38%) developed metastasis, 17 after discontinuing therapy, and 1 patient was alive at the end of the follow-up period. Metastasis involving the liver, the most common site of metastasis in uveal melanoma, occurred in 90% of IFN-treated patients and 86% of control patients who developed metastasis. The proportion of patients who died of metastatic disease was similar in the 2 groups: 30.6% (n ⫽ 37) in the IFN-treated group and 26.9% (n ⫽ 65) in the proton therapy or enucleation only group. Kaplan–Meier survival curves for melanoma-related mortality in IFN-treated patients and controls are shown in Figure 1. The 5-year melanoma-related death rates were 16.9% (95% confidence interval, 12.7%–22.4%) in the radiation or enucleation only group and 23.8% (95% confidence interval, 17.1%–32.6%) in the adjuvant therapy group. No material differences were observed with longer follow-up. Survival distributions were not significantly different after adjusting for ciliary body involvement (stratified log-rank test, P ⫽ 0.83), with cumulative death rates virtually identical regardless of tumor location (not shown). Results of multivariate Cox regression are shown in Table 3. Adjusting for known prognostic factors (LTD, age, ciliary body involvement, and extrascleral extension), rate ratios for melanoma-specific mortality in the “intent to treat” analysis (comparing patients assigned to receive IFN therapy after primary therapy with patients who had received primary therapy only) were not significantly different from the null (1.0). We completed further analyses evaluating the effect of the total IFN dose received (percent of the theoretic dose). We categorized subjects according to whether or not they completed the full 2-year IFN treatment regimen and compared these groups with their matched controls. When compared with no IFN treatment (historical controls), failure to complete the IFN regimen, defined as receiving ⱕ75% of the theoretic dose, was associated with a nonsignificant 40% increase in death rates, whereas successful completion of the target IFN dose, defined as having received ⬎75% of the theoretic dose, was associated with a nonsignificant 20% reduction in risk (Table 3). This pattern is observed in the Kaplan–Meier curves (Fig 2): Cumulative death rates were similar in patients who completed IFN and controls at both 5 years (17% in the radiation only group compared with 15% in those who

Lane et al 䡠 Adjuvant Interferon Therapy for Uveal Melanoma Table 3. Cox Regression Analysis of Adjuvant Interferon Therapy as a Predictor of Melanoma-related Mortality Multivariate Model†

CB-adjusted Model* Interferon Status Intent to treat None Any Theoretic dose None ⬎0% to ⱕ75% ⬎75%

No. of Deaths

RR

95% CI

P

RR

95% CI

P

65 37

ref 1.05

0.71–1.5

0.81

ref 1.02

0.67–1.5

0.91

65 19 18

ref 1.4 0.84

0.81–2.3 0.51–1.4

0.24 0.48

ref 1.4 0.82

0.76–2.3 0.49–1.3

0.31 0.44

CB ⫽ ciliary body; CI ⫽ confidence interval; LTD ⫽ largest tumor diameter; RR ⫽ rate ratio. *Treatment effect adjusted for CB involvement. Treatment effect adjusted for LTD, age at diagnosis/primary treatment, CB involvement, and extrascleral extension.

†

completed the entire IFN course) and 9 years (⬃28% in both groups) after completion of proton therapy or enucleation. However, higher cumulative mortality was observed in the group who discontinued IFN therapy (5- and 10-year rates ⱖ35%). This pattern was attenuated after excluding dropouts because of metastasis (Fig 3). The 5-year cumulative death rate was 26% in the group who discontinued IFN for reasons other than metastasis, higher than the groups who received primary therapy only or completed IFN therapy. However, by 9 years after primary therapy, melanoma mortality rates in this group were similar to those of the other treatment groups. We next considered whether timing of the initiation of adjuvant therapy with respect to diagnosis of the tumor influenced IFN efficacy, modeling treatment as a time-varying factor, and found no evidence that IFN affected survival (data not shown). These analyses had limited power because most patients (80%) began IFN treatment immediately after the tumor diagnosis. No material differences were observed for all-cause mortality (Fig 4). Death rates were slightly lower in the IFN group than in the control group after adjustment for other prognostic factors, but the difference was not statistically significant (rate ratio, 0.84; 95% confidence interval, 0.58 –1.2; P ⫽ 0.34).

Discussion In the present analysis, low-dose IFN therapy (3 MIU) did not reduce metastatic death rates in patients at increased risk of dying of uveal melanoma. We evaluated IFN-alfa-2a in

Figure 2. Unadjusted rates of melanoma-related mortality by duration of IFN treatment. IFN ⫽ interferon; PBI ⫽ proton beam irradiation.

this nonrandomized trial because of early reports from several small studies demonstrating a potential anti-tumor effect of IFN-alfa-2a in the treatment of metastatic cutaneous melanoma (⬃10%18 to 20%19,20 response rate). In addition, early results from a then ongoing World Health Organizationsponsored trial in Europe suggested prolonged survival in patients with cutaneous melanoma treated with adjunctive IFN-alfa-2a.21 Subsequently, the 2b form of IFN-alfa, which has similar immunomodulatory and anti-angiogenic actions, was also demonstrated to be effective for treatment of high-risk resected cutaneous melanoma. Failure to demonstrate any benefit associated with low-dose IFN therapy in the present study for patients with uveal melanoma, if not due to chance or bias, could possibly reflect biological differences in the pathogenesis of cutaneous and ocular melanomas.22–26 Overexpression of the c-myc oncogene, a cell-cycle regulator that is associated with poor prognosis in cutaneous melanoma27,28 but seems to be protective in uveal melanoma,29 may be of particular relevance with regard to IFN treatment. Tulley and colleagues30 demonstrated reduced sensitivity to IFN-␣-2b and reduced inhibition of tumor cell growth after introduction of IFN in uveal melanomas exhibiting high c-myc expression. These findings suggest that IFN should be more effective in patients with a poorer prognosis through low c-myc expression. In a murine model of metastatic

Figure 3. Unadjusted rates of melanoma-related mortality by duration of IFN treatment, excluding patients who discontinued IFN because of metastasis. IFN ⫽ interferon; PBI ⫽ proton beam irradiation.

2209

Ophthalmology Volume 116, Number 11, November 2009

Figure 4. Unadjusted rates of all-cause mortality by treatment group. IFN ⫽ interferon; PBI ⫽ proton beam irradiation.

ocular melanoma, an increase in natural killer cells after IFN treatment was associated with a decrease in micrometastases, and the magnitude of the effect was dependent on expression of MHC class I molecules on the tumor cells.31 Additional research is necessary to determine whether adjunctive IFN therapy may improve survival times in selected subgroups of patients. It is possible that the IFN dose (3 MIU TIW for 2 years) in the current study was too low to be efficacious. Patients with cutaneous melanoma (American Joint Committee on Cancer Classification stage IIB or stage III disease) receive an intravenous induction dose of 20 MIU/m2 IFN-alfa-2b every day for 1 month followed by a subcutaneous maintenance dose of 10 MIU/m2 TIW for 48 weeks, a regimen that was shown to extend both recurrence-free survival and overall survival in the Eastern Collaborative Oncology Group Study EST 1684.17 This is more than twice the dose administered in the current study. Findings in a subsequent study conducted by the Eastern Collaborative Oncology Group32 evaluating both low-dose and high-dose IFNalfa-2b in cutaneous melanoma demonstrated a reduction in recurrence rates only among patients receiving the higher dose. A prolonged treatment course with maintenance of threshold levels of IFN may be needed to destroy metastatic cells. Consistent with this possibility, patients in the present study who completed the full 2-year course of IFN therapy (45%) had a nonsignificant 40% reduction in risk of metastatic death. A small study of early-stage (American Joint Committee on Cancer Classification stage I or II) cutaneous melanoma, in which all patients were able to complete a low-dose adjuvant regimen of 3 MIU TIW given for 3 years, demonstrated a significant survival advantage in the IFNtreated patients.33 We hypothesized that early intervention with IFN, that is, before clinically detectable metastases develop and when tumor burden is small, would reduce the rate of melanoma progression, supported in a murine model of metastatic ocular melanoma.31,34 However, we found no benefit of IFN therapy introduced at the time of tumor diagnosis and initial treatment. Modeling of tumor doubling times suggests that micrometastases may develop as many as 5 years before the

2210

diagnosis of the primary tumor.35 Introduction of adjuvant IFN at the time of diagnosis may be suboptimal if metastatic cells are able to develop mechanisms to evade the immunomodulatory effects of IFN. Other possible reasons for failure of IFN therapy include the development of anti-IFN antibodies, which neutralizes the effect of exogenous IFN. Interferon antibodies have been detected in patients treated with IFN-␣-2a for a number of cancers,18,19,36,37 although in this study we did not detect anti-IFN antibodies in a subgroup of patients for whom enzyme-linked immunosorbent assays were performed in samples collected at baseline and 18 months posttreatment (Wee R, Lane AM, Gragoudas ES, Young LH, et al. Invest Ophthalmol Vis Sci 2006;47:ARVO E-Abstract 2253). As has been demonstrated in other studies, antibodies may develop earlier in the treatment regimen18,19 or peak and then decline,37,38 which could explain our negative results. However, none of the investigations with positive findings were able to establish an association between developing neutralizing antibodies and response to treatment. Continued research is needed to determine whether development of anti-IFN antibodies may predict IFN treatment response. Beyond the question of efficacy, there may be barriers to the widespread use of IFN because of its toxicity level. IFN therapy is not suitable for many patients; approximately one quarter of patients considered for inclusion in the current study had a medical condition for which IFN therapy was contraindicated. In addition, many of the patients who were able to start therapy were unable to tolerate the adverse effects of IFN and discontinued therapy, either by choice or because it was medically necessary. Patients were also older, on average (median age was 63 years), than patients enrolled in cutaneous melanoma trials (median age was 49 years),39 which may have contributed to relatively high rates of dropout and poor performance of IFN in this study relative to therapeutic trials in cutaneous melanoma. The high dropout rate in our study is not unlike that of other studies of IFN. However, we did not anticipate the observed levels of toxicity in our cohort. Compared with subjects treated in many previous studies who have already developed metastasis, our population was presumably healthier (given that they were free of metastasis at study entry) and their dosing regimen was much lower. These results indicate that evaluating efficacy of IFN at higher doses may not be feasible. In conclusion, despite the negative results in the current study, further adjuvant or neoadjuvant therapy trials using IFN regimens may be warranted. Yang and Grossniklaus40 experimentally demonstrated fewer hepatic micrometastases after treatment with combination IFN and angiostatin therapy than with either agent alone. Promising results have also been demonstrated in experiments using IFN-beta adenovirus vectors; this approach has the advantage of sustaining high circulating levels of IFN with a vector that has a predilection to transduce hepatocytes.41 Identifying reliable serum or tumor biomarkers associated with treatment response may also help to establish a role for IFN in a subgroup of patients.

Lane et al 䡠 Adjuvant Interferon Therapy for Uveal Melanoma

References 17. 1. Gragoudas E, Li W, Goitein M, et al. Evidence-based estimates of outcome in patients irradiated for intraocular melanoma. Arch Ophthalmol 2002;120:1665–71. 2. Sparano JA, Lipsitz S, Wadler S, et al. Phase II trial of prolonged continuous infusion of 5-fluorouracil and interferonalpha in patients with advanced pancreatic cancer: Eastern Cooperative Oncology Group Protocol 3292. Am J Clin Oncol 1996;19:546 –51. 3. Escudier B, Pluzanska A, Koralewski P, et al., AVOREN Trial Investigators. Bevacizumab plus interferon alfa-2a for treatment of metastatic renal cell carcinoma: a randomised, doubleblind phase III trial. Lancet 2007;370:2103–11. 4. Hudes G, Carducci M, Tomczak P, et al., Global ARCC Trial. Temsirolimus, interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med 2007;356:2271– 81. 5. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 2007;356:115–24. 6. Pyrhonen S, Salminen E, Ruutu M, et al. Prospective randomized trial of interferon alfa-2a plus vinblastine versus vinblastine alone in patients with advanced renal cell cancer. J Clin Oncol 1999;17:2859 – 67. 7. Cascinelli N, Belli F, MacKie RM, et al. Effect of long-term adjuvant therapy with interferon alpha-2a in patients with regional node metastases from cutaneous melanoma: a randomised trial. Lancet 2001;358:866 –9. 8. Atzpodien J, Lopez Hanninen E, Kirchner H, et al. Chemoimmunotherapy of advanced malignant melanoma: sequential administration of subcutaneous interleukin-2 and interferonalpha after intravenous dacarbazine and carboplatin or intravenous dacarbazine, cisplatin, carmustine and tamoxifen. Eur J Cancer 1995;31A:876 – 81. 9. Groenewegen G, Bloem A, De Gast GC. Phase I/II study of sequential chemoimmunotherapy (SCIT) for metastatic melanoma: outpatient treatment with dacarbazine, granulocytemacrophage colony-stimulating factor, low-dose interleukin-2, and interferon-alpha. Cancer Immunol Immunother 2002;51: 630 – 6. 10. Becker JC, Terheyden P, Kampgen E, et al. Treatment of disseminated ocular melanoma with sequential fotemustine, interferon alpha, and interleukin 2. Br J Cancer 2002;87: 840 –5. 11. Kivela T, Suciu S, Hansson J, et al. Bleomycin, vincristine, lomustine and dacarbazine (BOLD) in combination with recombinant interferon alpha-2b for metastatic uveal melanoma. Eur J Cancer 2003;39:1115–20. 12. Nathan FE, Berd D, Sato T, et al. BOLD⫹interferon in the treatment of metastatic uveal melanoma: first report of active systemic therapy. J Exp Clin Cancer Res 1997;16:201– 8. 13. Pyrhonen S, Hahka-Kemppinen M, Muhonen T, et al. Chemoimmunotherapy with bleomycin, vincristine, lomustine, dacarbazine (BOLD), and human leukocyte interferon for metastatic uveal melanoma. Cancer 2002;95:2366 –72. 14. Solti M, Berd D, Mastrangelo MJ, Sato T. A pilot study of low-dose thalidomide and interferon alpha-2b in patients with metastatic melanoma who failed prior treatment. Melanoma Res 2007;17:225–31. 15. de Waard-Siebinga I, Creyghton WM, Kool J, Jager MJ. Effects of interferon alfa and gamma on human uveal melanoma cells in vitro. Br J Ophthalmol 1995;79:847–55. 16. Grob JJ, Dreno B, de la Salmoniere P, et al; French Cooperative Group on Melanoma. Randomised trial of interferon alpha-2a as adjuvant therapy in resected primary melanoma

18.

19. 20. 21. 22.

23. 24. 25. 26.

27. 28.

29. 30. 31.

32.

33.

34.

35.

thicker than 1.5 mm without clinically detectable node metastases. Lancet 1998;351:1905–10. Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996;14:7–17. Legha SS, Papadopoulos NE, Plager C, et al. Clinical evaluation of recombinant interferon alfa-2a (Roferon-A) in metastatic melanoma using two different schedules. J Clin Oncol 1987;5:1240 – 6. Creagan ET, Ahmann DL, Green SJ, et al. Phase II study of low-dose recombinant leukocyte A interferon in disseminated malignant melanoma. J Clin Oncol 1984;2:1002–5. Creagan ET, Ahmann DL, Green SJ, et al. Phase II study of recombinant leukocyte A interferon (rIFN-alpha A) in disseminated malignant melanoma. Cancer 1984;54:2844 –9. Cascinelli N, Bufalino R, Morabito A, Mackie R. Results of adjuvant interferon study in WHO melanoma programme [letter]. Lancet 1994;343:913– 4. Sisley K, Cottam DW, Rennie IG, et al. Non-random abnormalities of chromosomes 3, 6, and 8 associated with posterior uveal melanoma. Genes Chromosomes Cancer 1992;5:197– 200. Edmunds SC, Cree IA, Di Nicolantonio F, et al. Absence of BRAF gene mutations in uveal melanomas in contrast to cutaneous melanomas. Br J Cancer 2003;88:1403–5. Monzon J, Liu L, Brill H, et al. CDKN2A mutations in multiple primary melanomas. N Engl J Med 1998;338:879 – 87. Brantley MA Jr, Harbour JW. Deregulation of the Rb and p53 pathways in uveal melanoma. Am J Pathol 2000;157:1795– 801. Edmunds SC, Kelsell DP, Hungerford JL, Cree IA. Mutational analysis of selected genes in the TGFbeta, Wnt, pRb, and p53 pathways in primary uveal melanoma. Invest Ophthalmol Vis Sci 2002;43:2845–51. Ross DA, Wilson GD. Expression of c-myc oncoprotein represents a new prognostic marker in cutaneous melanoma. Br J Surg 1998;85:46 –51. Grover R, Ross DA, Wilson GD, Sanders R. Measurement of c-myc oncoprotein provides an independent prognostic marker for regional metastatic melanoma. Br J Plast Surg 1997;50: 478 – 82. Chana JS, Wilson GD, Cree IA, et al. c-myc, p53, and Bcl-2 expression and clinical outcome in uveal melanoma. Br J Ophthalmol 1999;83:110 – 4. Tulley PN, Neale M, Jackson D, et al. The relation between c-myc expression and interferon sensitivity in uveal melanoma. Br J Ophthalmol 2004;88:1563–7. Yang H, Dithmar S, Grossniklaus HE. Interferon alpha 2b decreases hepatic micrometastasis in a murine model of ocular melanoma by activation of intrinsic hepatic natural killer cells. Invest Ophthalmol Vis Sci 2004;45:2056 – 64. Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol 2000;18:2444 –58. Rusciani L, Petraglia S, Alotto M, et al. Postsurgical adjuvant therapy for melanoma: evaluation of a 3-year randomized trial with recombinant interferon-alpha after 3 and 5 years of follow-up. Cancer 1997;79:2354 – 60. Dithmar S, Rusciano D, Lynn MJ, et al. Neoadjuvant interferon alfa-2b treatment in a murine model for metastatic ocular melanoma: a preliminary study. Arch Ophthalmol 2000;118: 1085–9. Eskelin S, Pyrhonen S, Summanen P, et al. Tumor doubling times in metastatic malignant melanoma of the uvea: tumor

2211

Ophthalmology Volume 116, Number 11, November 2009 progression before and after treatment. Ophthalmology 2000; 107:1443–9. 36. Rajan GP, Seifert B, Prummer O, et al. Incidence and in-vivo relevance of anti-interferon antibodies during treatment of low-grade cutaneous T-cell lymphomas with interferon alpha-2a combined with acitretin or PUVA. Arch Dermatol Res 1996; 288:543– 8. 37. Prummer O, Delta-P Study Group. Interferon-alpha antibodies in patients with renal cell carcinoma treated with recombinant interferon-alpha-2A in an adjuvant multicenter trial. Cancer 1993;71:1828 –34. 38. Ronnblom LE, Janson ET, Perers A, et al. Characterization of anti-interferon-alpha antibodies appearing during recombinant

interferon-alpha 2a treatment. Clin Exp Immunol 1992;89: 330 –5. 39. Kirkwood JM, Manola J, Ibrahim J, et al. A pooled analysis of Eastern Cooperative Oncology group and intergroup trials of adjuvant high-dose interferon for melanoma. Clin Cancer Res 2004;10:1670 –7. 40. Yang H, Grossniklaus HE. Combined immunologic and anti-angiogenic therapy reduces hepatic micrometastases in a murine ocular melanoma model. Curr Eye Res 2006;31: 557– 62. 41. Alizadeh H, Howard K, Mellon J, et al. Reduction of liver metastasis of intraocular melanoma by interferon-beta gene transfer. Invest Ophthalmol Vis Sci 2003;44:3042–51.

Footnotes and Financial Disclosures Originally received: December 17, 2008. Final revision: March 23, 2009. Accepted: April 24, 2009. Available online: September 10, 2009.

Society annual meeting, September 18 to 21, 2003, Chicago, Illinois, and the Aegean Retina VIII meeting, July 11 to 13, 2003, Santorini, Greece. Manuscript no. 2008-1513.

1

Retina Service, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts. 2

Moffitt Cancer Center, Tampa, Florida.

3

Department of Hematology-Oncology, Massachusetts General Hospital, Boston, Massachusetts. 4

Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts. Parts of this paper were presented at the 26th annual Macula Society meeting, February 26 to March 1, 2003, Naples, Florida, the 2003 Retina

2212

Financial Disclosure(s): Evangelos Gragoudas is a paid consultant for and has an equity ownership in Jerini Ophthalmics. This research was conducted with funding from Roche Laboratories, Nutley, New Jersey, and the Massachusetts Eye and Ear Infirmary Melanoma Research Fund. The funding organization approved the design of the study and the study manuscript. No other authors have any financial interests to disclose. Correspondence: Anne Marie Lane, MPH, Retina Service, Massachusetts Eye and Ear Infirmary, 243 Charles St., Boston, MA 02114. E-mail: alane@meei. harvard.edu.