GM-CSF and IL-2 Combination as Adjuvant Therapy in Cutaneous Melanoma

Cytokines have been used in the treatment of patients with cutaneous melanoma. Granulocyte-macrophage colony-stimulating factor (GM-CSF, sargramostim [Leukine]) leads to dendritic cell/macrophage priming and activation, and also increases interleukin-2 (IL-2) receptor expression on T lymphocytes. IL-2 creates lymphokineactivated killer cells and tumor-infiltrating lymphocyte cells. In this open-label, single-arm study of 16 high-risk patients, we combined these two agents to take advantage of their different but complementary functions. All patients underwent potentially curative surgery. Postoperatively, each patient received GM-CSF at 125 μg/m²/d subcutaneously (SC) for 14 days; this was followed by IL-2 at 9 million IU/m²/d SC for 4 days, and then 10 to 12 days of no treatment. In addition, patients who had large tumors that could yield over 100 million live tumor cells received autologous melanoma vaccines. The duration of follow-up ranged from 21 to 42 months (median: 27 months). During follow-up, five patients developed metastases. This program was carried out on an outpatient basis, and no hospitalization was required. It was well tolerated with minimal side effects. The combination treatment regimen of GM-CSF and IL-2 with or without autologous vaccine used adjuvantly appears to benefit high-risk melanoma patients; further clinical testing of this regimen is warranted.

Several cytokines have been utilized in the management of patients with cutaneous melanoma. Interferon alfa-2b (IFN) is the only agent approved by the US Food and Drug Administration (FDA) as adjuvant therapy in melanoma. However, it has had borderline survival benefits that disappear during follow-up. Interferon alfa-2b has to be administered for 1 year at a high dose that is associated with a high toxicity level, and it cannot be administered to the elderly. In 2002, the Oncology Advisory Committee of the FDA recommended that physicians do not have to offer IFN to patients as it is no longer considered standard treatment.

On the other hand, interleukin-2 (IL-2) is approved by the FDA for the management of metastatic melanoma. The in vitro culturing of peripheral blood mononucleated cells with IL-2 results in lymphokine-activated killer (LAK) cells. In addition, culturing lymphocytes from metastatic melanoma, specifically from soft tissues, with IL-2 gives rise to tumor-infiltrating lymphocyte (TIL) cells, which are specific cytotoxic lymphocytes. These TIL cells are more potent than the LAK cells.

The newest cytokine to be used as adjuvant therapy in melanoma is granulocyte- macrophage colony-stimulating factor (GM-CSF, sargramostim [Leukine]). It is the least toxic cytokine, and early results from its use as adjuvant therapy in melanoma are very encouraging. GM-CSF serves as the principal mediator for the proliferation, maturation, and migration of dendritic cells, which are antigenpresenting cells (APCs) that play a major role in the induction of primary and secondary T-cell immune responses.[1,2]

Spitler and coworkers evaluated the safety and efficacy of adjuvant GMCSF in patients with stage III and resected stage IV melanoma who were at a very high risk of recurrence and metastases. Their results showed significant improvement in survival when compared to historical controls.[3] While nearly all immune responses require T lymphocytes, appropriate presentation of tumor antigen with costimulatory molecules and cytokines is necessary for triggering an immune response. Antigen-presenting cells are crucial players in this process, as they are phagocytic cells that trap, process, and display antigens for T-cell recognition.

In this study we utilized two cytokines, GM-CSF and IL-2, each having a function different from but complementary to that of the other. In addition, when GM-CSF was injected directly at melanoma sites, it was noted that the absolute number of CD4+ T lymphocytes and the number of dendritic cells in the epidermis were increased. Furthermore, the IL-2 receptor expression on T lymphocytes, both around and within the tumor, was increased. Such findings correlated with the responses seen in a study by Si and associates.[4] This further justifies the use of the combination of GM-CSF and IL-2.

Materials and Methods

This was an open-label, single-arm study, approved by the Institutional Review Board. The high-risk patients who participated in this study underwent potentially curative surgery, eliminating all gross tumor. The study was comprised of patients with primary tumors of 2 to 4 mm deeply invasive melanoma with ulceration, patients with > 4 mm deeply invasive melanoma with and without ulceration, those with metastases to their regional lymph nodes, and patients with resected distant metastases with no gross residual tumor.

FIGURE 1

Schema-Open-label single-arm study of GM-CSF and IL-2 in 16 patients with cutaneous melanoma

Postoperatively, each patient received GM-CSF at 125 μg/m²/d subcutaneously (SC) for 14 days. This was followed by IL-2 at 9 million IU/m²/d SC for 4 days. The patient then received no treatment for the next 10 to 12 days (see Figure 1). Each of the first 16 patients received six cycles over a 6-month period. The protocol was then modified to a 2-year program; to date, 12 patients have been enrolled in the modified protocol. 

In addition, patients who had large tumors that could yield over 100 million live tumor cells received autologous melanoma vaccines. All patients were observed closely without further therapy. In the event of recurrence or metastases, the patient was evaluated for resection; if surgery rendered the patient grossly disease-free, he or she was treated on the same program.

Results

The first 16 patients enrolled in this study, who received only 6 months of adjuvant GM-CSF and IL-2, included 5 patients with 2 to 4 mm deeply invasive melanoma with ulceration, 3 patients with > 4 mm deeply invasive melanoma (2 of these patients had ulcerated lesions), 5 patients with regional lymph node metastases, and 3 patients with resected distant metastases.

There were eight men and eight women, ranging in age from 33 to 75 years (median: 50 years). The duration of follow-up ranged from 21 to 42 months (median: 27 months). During follow-up, five patients developed metastases. Two patients underwent resection of their metastases (one from lung and the other from soft tissue and adrenal gland) and are being treated on the 2-year program. These two patients are living free of disease at 1 and 2 years since the initiation of their 2-year treatment. The third patient died of metastases, and another patient is living with metastases. The fifth patient had brain metastasis that was resected, and received postoperative brain irradiation, and is alive with no evidence of disease for 12 months.

Toxicity

This program was carried out on an outpatient basis, and no hospitalization was required. It was well tolerated with minimal side effects, and we felt it was safe to extend it to 2 years.

GM-CSF-Related Toxicity

After administration of GM-CSF, five patients experienced intermittent sternal chest pain of 10 to 15 minutes duration, which was relieved with oral acetaminophen administration (1 g). There were no electrocardiogram or cardiac enzyme changes. Such chest pain did not recur with repeated administration of GM-CSF.

One patient developed clinical jaundice with total serum bilirubin of 3.6 mg/dL, aspartate aminotransferase 84 IU/L, alanine aminotransferase 190 IU/L, and alkaline phosphatase 267 IU/L. GM-CSF was discontinued and all the laboratory values returned to normal within 11 days after cessation of GM-CSF.

Five patients developed reactions at the injection sites in the form of itching, induration, erythema, and warmth (simulating cellulitis). Preicing the injection site for 1 to 2 minutes before GM-CSF, allowing the drug to warm up to room temperature before its administration, and topical application of diphenhydramine have greatly diminished such skin reactions.

Thirteen patients developed white cell counts of > 20,000/μL. This usually occurred at the end of the 14-day cycle and did not require any intervention. The white blood cell count returned to the normal range in less than 1 week.

Four patients experienced constitutional symptoms in the form of mild diarrhea, fever, chills, fatigue, headache, general muscle ache, and (rarely) vomiting. All these were grade 1, based on the National Cancer Institute Common Toxicity Criteria.

IL-2-Related Toxicity

After administration of IL-2, one patient experienced chest pain that lasted for about 15 minutes and resolved spontaneously. There was no evidence of cardiopulmonary disease. The pain did not recur with further IL-2 administration.

Four patients experienced water retention. Each gained 5 to 7 pounds during different cycles of IL-2. They presented with swelling of the hands, feet, and face and were managed with varying doses of furosemide.

Twelve patients experienced more than one form of dermal toxicity at random points while receiving the treatment. The most common toxicity was flushing, which resolved spontaneously without any intervention several days after completion of therapy. Peeling of the skin of the palms and maculopapular rash on the torso occurred rarely and were managed by skin moisturizers and diphenhydramine. These symptoms all resolved when patients were not taking the drug.

Nine patients experienced diarrhea. One patient had grade 3 diarrhea, based on frequency, which required a 50% dose reduction. This was managed by administration of oral fluids, and the symptoms resolved at the completion of therapy.

Subcutaneous nodules developed under the skin at the injection sites in all patients and resolved in 1 to 2 months postinjection. Fourteen patients developed constitutional symptoms including fever, chills, fatigue, headache, muscle aches, and flu-like symptoms. These were managed with oral acetaminophen and diphenhydramine. Some patients received meperidine hydrochloride for chills. All these symptoms were transient grade 1 to 2, and resolved at the completion of the drug regimen. 

Two patients developed subcutaneous nodules away from any GMCSF or IL-2 injection sites. One of these patients had two subcutaneous nodules: one on the left chest wall and the other on the right. The other patient had a subcutaneous nodule on his right chest wall. All three nodules were excised. The pathologic examination revealed them to be lymph nodes packed with T lymphocytes with few foci of B cells (Figure 2).

Conclusions

FIGURE 2

Excised Subcutaneous Nodule-Pathologic examination of three excised subcutaneous nodules that developed away from injection sites in two patients revealed them to be lymph nodes packed with T lymphocytes with few foci of B cells

The goal of adjuvant therapy after surgery for malignant melanoma is the suppression of progressive disease caused by micrometastases. High-risk patients, such as those in this trial, often require adjuvant therapy for the successful treatment of their disease. Overall, however, results from trials of adjuvant therapy for melanoma remain disappointing.

No standard treatment to prevent recurrence of melanoma exists. Researchers employing innovative treatment strategies, most centering on the use of immunotherapy, are seeking to improve the prognosis for melanoma patients. Interferon alfa has demonstrated statistically significant improvements in disease-free and overall survival, but trial results have been inconsistent, and the role of this agent as adjuvant treatment of stage II/III melanoma is not well defined.[5-8] The adjuvant use of other biologic agents has been widely studied. Interleukin- 2 has been shown to be effective, at higher doses, in advanced metastatic melanoma.[9-11] However, high-dose IL-2 can result in significant toxicity.[12] Vaccines designed to induce active antitumor immune responses have shown promise and continue to be tested, either alone or in combination with other immunotherapeutics.[ 13-16]

Macrophage-derived factors have been shown to be responsible for the induction of angiostatin production.[ 17] Upregulation of murine metalloelastase expression in macrophages by GM-CSF produced by local tumors can lead to generation of angiostatin and, thereby, the growth suppression of distant metastases.[18]

In the trial conducted by Spitler et al, the cytokine GM-CSF was given at a dosage of 125 μg/m² for 2 weeks, followed by 2 weeks off therapy. Treatment was continued until recurrence or until the patient had been tumor-free for 3 years. The regimen was well tolerated, with an acceptable toxicity profile. Major adverse events were injection site reactions and grade 1 malaise, fatigue, and arthralgias. The preliminary efficacy results of this trial, suggesting a survival benefit for the patients treated with GM-CSF (P =.0005), are similarly encouraging.[3]

A later pilot trial assessed the safety of adjuvant, daily GM-CSF, at a dosage of 150 μg/d SC for 2 years, in high-risk patients with resected melanoma. At a median time on treatment of 1 year, toxicity was mild, consisting of injection site reactions and bone pain that was controlled with antiinflammatory medication. Leukocytosis peaked at a median of 23,500/μL (range: 16,200-60,000/μL) at a median of 3 months of treatment, but it was not associated with clinical complications. Relapse-free survival at 1 year was 88.8%.[19]

Our results lend further support to clinical trials of GM-CSF in the treatment of advanced malignant melanoma. The combination treatment regimen of GM-CSF and IL-2 with or without autologous vaccine used adjuvantly appears to benefit high-risk melanoma patients; further clinical testing of this regimen is warranted. The GM-CSF/IL-2 regimen was feasible given on an outpatient basis, and was associated with relatively low toxicity. Adverse events of the combination, at this early assessment, were minimal and manageable. In an attempt to intensify this therapy without increasing the dosage and the likelihood of leukocytosis, this treatment protocol has been extended to 2 years.

Financial Disclosure:Dr. Elias has received grants and/or research support from the Russell Fund. He receives or has received other fiancial or material support from Chiron Corporation and Berlex Laboratories. Chiron provided IL-2 at no cost to patients on this study.

References:

1. Szabolcs P, Avigan D, Gezelter S, et al: Dendritic cells and macrophages can mature independently from a human bone marrow-derived, post-colony-forming unitintermediate. Blood 87: 4520-4530, 1996.
2. Armitage JO: Emerging applications of recombinant human granulocyte-macrophage colony stimulating factor. Blood 92: 4491- 4508, 1998.
3. Spitler LE, Grossbard ML, Emstoff, MS, et al: Adjuvant therapy of stage III and IV malignant melanoma using granulocyte-macrophage colony stimulating factor. J Clin Oncol 18:1614-1621, 2000.
4. Si Z, Hershey P, Coates AS: Clinical responses and lymphoid infiltrates in metastatic melanoma following treatment with intralesional GM-CSF. Melanoma Research 6:247- 255, 1996.
5. Creagan ET, Ahmann D L, Frytak S, et al: Three consecutive phase II studies of recombinant interferon α 2a in advanced malignant melanoma. Cancer 59:638-646, 1987.
6. Hernberg M, Pyrhonen S, Muhonen T, et al: Regimens with or without interferon-α as treatment for metastatic melanoma and renal cell carcinoma: An overview of randomized trials. J Immunother 22:145-154, 1999.
7. Kirkwood JM, Strawderman MH, Ernstoff MS, et al: IFN adjuvant therapy of high risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 14:7-17, 1996.
8. Mitchell MS, Jakowatz J, Harel W, et al: Increased effectiveness of interferon α 2b following active specific immunotherapy for melanoma. J Clin Oncol 12: 402-411, 1994.
9. Lindsey KR, Rosenberg SA, Sherry RM, et al: Impact of the number of treatment courses on the clinical response of patients who receive high-dose bolus interleukin-2. J Clin Oncol 18:1954-1959, 2000.
10. Keilholz U, Conradt C, Legha SS, et al: Results of interleukin-2-based treatment in advanced melanoma: A case record-based analysis of 631 patients. J Clin Oncol 16:2921- 2929, 1998.
11. Rosenberg SA: Interleukin-2 and the development of immunotherapy for the treatment of patients with cancer. Cancer J Sci Am 6(suppl 1):S2–S7, 2000.
12. Lens MB: The role of biological response modifiers in malignant melanoma. Expert Opin Biol Ther 3:1225-1231, 2003.
13. Slingluff CL Jr, Petroni GR, Yamshchikov GV, et al: Immunologic and clinical outcomes of vaccination with a multiepitope melanoma peptide vaccine plus low-dose interleukin-2 administered either concurrently or on a delayed schedule. J Clin Oncol 15:4474-4485, 2004.
14. Trefzer U, Herberth G, Wohlan K, et al: Vaccination with hybrids of tumor and dendritic cells induces tumor-specific T-cell and clinical responses in melanoma stage III and IV patients. Int J Cancer 110:730-740, 2004.
15. Vilella R, Benitez D, Mila J, et al: Pilot study of treatment of biochemotherapy-refractory stage IV melanoma patients with autologous dendritic cells pulsed with a heterologous melanoma cell line lysate. Cancer Immunol Immunother 53:651-658, 2004.
16. Haenssle HA, Krause SW, Emmert S, et al: Hybrid cell vaccination in metastatic melanoma: Clinical and immunologic results of a phase I/II study. J Immunother 27:147-155, 2004.
17. Pinedo HM, De Gruijl TD, Van Der Wall E, et al: Biological concepts of prolonged neoadjuvant treatment plus GM-CSF in locally advanced tumors. The Oncologist 5:497-500, 2000.
18. Dong Z, Yoneda J, Kumar R, et al: Angiostatin-mediated suppression of cancer metastases by primary neoplasms engineered to produce granulocyte/macrophage colonystimulating factor. J Exp Med 188:755-763, 1998.
19. Isla D, Filipovich E, Mayordomo JI, et al: Daily GM-CSF for patients with very highrisk resected melanoma: A pilot trial (abstract 2784). Proc Am Soc Clin Oncol 21:241B, 2002.