1149 Phase II Trial of Subcutaneously Administered Granulocyte-Macrophage Colony-Stimulating Factor in Patients with Metastatic Renal Cell Carcinoma Edward WOS, D.0.’ Thomas Olencki, D.O.’,~ Laurie Tuason, M.S? G. Thomas Budd, M.D.’ David Peereboom, M.D.”* Kate Sandstrom, A.N.’ Denise McLain, B.S? James Finke, P ~ . D . * , ~ Ronald M. Bukowski, M . D . ’ ~ ~ ~ ~ ’ Department of Hematology and Medical Oncology, The Cleveland Clinic Foundation, Cleveland. Ohio. Cancer Center Experimental Therapeutics Program, The Cleveland Clinic Foundation, Cleveland. Ohio. Department of Biostatistics and Epidemiology, The Cleveland Clinic Foundation, Cleveland, Ohio. Department of Immunology. The Cleveland Clinic Foundation, Cleveland, Ohio. This work was supported in part by a grant from the Schering Plough Corporation. Address for reprints: Ronald M. Bukowski, MI.,The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195. Received October 18,1995;accepted December a, 1995. 0 1996 American Cancer Society BACKGROUND. Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a cytokine that is involved in the differentiation and proliferation of various hematopoietic precursors. It also has been reported to enhance the antitumor activity of various mature effector cells. Previous reports have noted preclinical antitumor activity in a murine model utilizing genetically engineered tumor cells and instances of tumor regression in patients with solid tumors receiving GM-CSF. In the present study, a Phase I1 trial of human recombinant GM-CSF (GM-CSF,,,) in patients with metastatic renal cell carcinoma (RCC) was conducted to investigate further the potential antitumor activity of this cytokine. METHODS. Twenty-six eligible patients with metastatic RCC received 3 pglkg of GM-CSF,,, subcutaneously for 14 days, with cycles repeated every 28 days. RESULTS. Two of 26 patients (8%; 95% confidence interval 1-25%) demonstrated partial tumor responses during GM-CSF,,, therapy. Both individuals who responded had received prior therapy. A median of three cycles per patient were administered, and toxicity was mild. CONCLUSIONS. GM-CSFrhmay mediate tumor regression in patients with metastatic RCC; however, the level of activity in previously treated patients is low. Cancer 1996; 7 2 1 149-53. 0 1996 American Cancer Society. KEYWORDS: renal cell carcinoma, metastatic, GM-CSF, Phase II. G ranulocyte-macrophage colony stimulating factor (GM-CSF)is a cytokine that is responsible in part for differentiation and proliferation of various hematopoietic precursors. Specifically, GM-CSF stimulates clonal growth of CFU-GEMM,’ promotes granulocyte and macrophage differentiation and proliferation,* stimulates erythroid burst activity,‘ and may initiate eosinophil and megakaryocytic differentiation.’ It is secreted primarily by activated CD4’ lymphocytes but also is produced by a wide variety of cells including CD8’, B lymphocytes, endothelial cells, fibroblasts, and mast Human CM-CSF is a homodimeric glycoprotein consisting of two identical protein chains linked by disulfide bonds.’ When it is administered to either hematologically normal or leukopenic cancer patients, there is a reversible, dose-dependent leukocytosis consisting of functionally normal granulocytes, monocytes, and eosin~phils.~-” In addition, this cytokine can enhance various activities of mononuclear effector cells. In vitro studies have demonstrated that GM-CSF-treated monocytes display increased adherence, increased production of superoxides, interferon (IFN)a, interleukin (IL)-lp,1L-6,’*-I4 and tumoricidal activity.I3.l5 In vitro cytotoxic effects of GM-CSF have been demonstrated on various human tumor cell lines,I6 including the U937 monoblast cell 1150 CANCER March 15,1996 I Volume 77 I Number 6 line.I7Finally, Dranoff et al.” have demonstrated in murine models that tumor cells genetically engineered to produce GM-CSF can stimulate development of specific antitumor immunity. Based on these findings and the enhanced in vitro effector functions found in our recent Phase I trial of GM-CSF in lung cancer patients,” a Phase I1 trial of GM-CSF in patients with renal cell carcinoma was initiated. This tumor was selected in view of reports documenting responses to various cytokines, such as IFNaL9and IL-2.” Our results demonstrate brief antitumor responses to systemically administered GM-CSF in 2 of 26 patients. MATERIALS AND METHODS Schering Plough Corporation (Kenilworth,NJ). GM-CSFm was supplied as a lyophilized powder formulated with mannitol, human serum albumin, polyethylene glycol, and citratelphosphate buffer. The vials each contained 400 pg or 700 pg and were reconstituted with 0.25-1.0 ml of sterile water. CM-CSFrhwas given subcutaneously at a dose of 3 pglkglday on days 1-14. Cycles were repeated every 4 weeks, unless excessive toxicity occurred or progressive disease developed. Dose escalation was not permitted. Patients who required corticosteroids, hormone therapy, chemotherapy, or radiation therapy during GM-CSF,h treatment were removed from the study. Patient Selection Patients were eligible if they had metastatic or surgically unresectable renal cell carcinoma with objectively measurable disease and no clinically evident central nervous system metastases. Eligibility criteria included histologically proven renal cell carcinoma with an Eastern Cooperative Oncology Group Performance Status of 0-2, age 2 18 years, and a minimum life expectancy of 3 months. Adequate hematopoietic, renal, and hepatic function were required and were defined as follows: white blood cell W C ) count 2 3,500lp1, hemoglobin (Hgb) 2 9.5 gmldl, platelet count 2 100,000lp1, serum creatinine s 1.8 mgldl, and a total bilirubin of 51.5 mgldl. Normal cardiovascular and pulmonary function were also required and were defined as the absence of a previous history of serious cardiac arrhythmia, myocardial infarction, or congestive heart failure, and a New York Heart Association classification of 4 1 . In addition, the absence of significant effusions or ascites and a carbon monoxide diffusing capacity D L o z 50% predicted were required. Prior therapy with cytokines was allowed, but patients who had received prior GM-CSF were excluded. One prior chemotherapy regimen or radiation therapy was allowed, but it must have been completed >28 days prior to the start of therapy, and prior surgery must have been completed > 14 days before study entry. Patients with recent infections requiring systemic antibiotics within 21 days prior to study entry were ineligible. Women of child-bearing age must have had a negative pregnancy test or be using a method of contraception. Informed consent, consistent with federal and institutional guidelines, was obtained from all patients. Exclusion criteria included history of malignancy other than basal cell carcinoma or carcinoma in situ of the cervix within the past 3 years, positive HIV or hepatitis serology, or a requirement for corticosteroids. Dose Modifications The NCI common clinical toxicity criteria was employed to assess side effects. In patients who developed Grade I11 toxicity (with the exception of a fever), treatment was withheld until there was a return to Grade I or normalization. GM-CSFrhwas then resumed at 1.5 pglkglday. Following reinstitution of treatment, recurrence of the identical Grade 111 toxicity required removal from the study. Development of Grade IV toxicity required discontinuatiOn O f GM-CSF,h. Dose modifications for leukocytosis were also employed. In patients with a total WBC = 75,000lp1, GMCSF,,, was withheld and was resumed at 1.5 pglkglday when the WBC decreased to <75,OOO/pl. Recurrence of leukocytosis after an initial dose reduction required a second 50% dose reduction of GM-CSFrh,and subsequent recurrence required removal from the study. Dose modifications for thrombocytosis were also developed, but they were not required during the trial. Treatment Human recombinant GM-CSF (GM-CSF,h) was obtained from the National Cancer Institute and supplied by the Statistical Considerations A standard Phase I1 trial design was employed. Initially, 14 eligible patients were entered. If no responses were Study Monitoring AU studies were performed within 14 days prior to study entry. These included a complete physical examination, biochemical studies [total protein, albumin, calcium, phosphorous, glucose, uric acid, total bilirubin, alkaline phosphatase, serum lactic dehydrogenase (LDH),serum glutamic oxaloacetic transaminase, sodium, potassium, chloride, C02, blood urea nitrogen, and creatininel, a complete blood count with differential, a chest x-ray, and imaging studies to assess measurable disease. A head CT scan was performed if it was clinically indicated. Oral temperatures were obtained at baseline and then daily (by the patient) during GM-CSF,h therapy. Complete blood counts and biochemical studies were obtained weekly, and tumor measurements were obtained prior to each cycle of GM-CSFrh.Previously reported response criteria and definitions were utilized.2’ Phase I1 Trial GM-CSF In Metastatic Renal Cell Carcinoma/Wos et al. observed, then a response rate of <20% at the 95% confidence level would be estimated. In the presence of one response, accrual was to continue until 25 evaluable patients had been treated. This allowed estimation of the true response with a standard error of at most 10%. Summary statistics were calculated for all biochemical and hematologic data. Response duration was defined as the initial time of documented response to time of progressive disease. RESULTS From February, 1991, through June, 1994, 28 patients were enrolled. Two patients were ineligible: one secondary to a positive hepatitis serology and the second because of excessive prior chemotherapy. The characteristics of the 26 eligible patients included a sex distribution of 10 females and 16 males, a median age of 58.5 years (range 31-80), and 22 patients with an Eastern Cooperative Oncology Group Performance Status of 0-1. The median performance status was 1, and 20 of 26 (77%) eligible patients had received prior therapy. This included rIL-2 c I F N a (75%),IL-6 or IL-4 (30%), chemotherapy (25%), or radiation therapy (15%).Of 26 patients, 19 (73%) had undergone a prior nephrectomy. Sites of metastases included lungs (77%), liver (15%), and soft tissuellymph nodes (23%). The median number of cycles of GM-CSF,,, administered to patients was three (range 1-11), and 2 of 26 patients had partial responses (response rate 8%; 95% confidence interval, 1-25%) during therapy. One patient was a 49-year-old male with lung metastases who had previously received IFNa. The second was a 31-year-old male with lung metastases who had previously received radiation therapy and the combination of fluorouracil, IL2, and IFNa. The response durations in these two patients were 1 and 2 months, respectively. The most common adverse events seen included mild-to-moderate arthralgias (69%), feverlchills (81%), and erythema and/or rash (69%).Several Grade 111-IV toxicities occurred during GM-CSFrhadministration, including neutropenia (two patients), elevated partial thromboplastin time (one patient), alkaline phosphatase (two patients), total bilirubin (two patients), serum glutamic pyruvic transaminase (one patient), as well as hypercalcemia (two patients) and hyperglycemia (one patient). These toxicities did not require dose modifications or postponement of GM-CSFrhadministration, and the majority were disease related. The hematologic effects of GM-CSF,,, were observed in all patients and included leukocytosis and neutrophilia. In addition, most patients had transient eosinophilia. During cycle 1, the median peak neutrophil level was 19,3OO/p1and occurred between days 11- 14. Monocytosis was also noted, with a median peak value of 1,250/ 1151 occurring between days 8- 19 (median day 15). Thrombocytosis associated with GM-CSFrhadministration was not seen. p1 DISCUSSION The potential antitumor effects of GM-CSF have been demonstrated both in vitro and in murine tumor models.” The present study was the first Phase I1 trial to address this issue in patients with metastatic malignancy. Our results demonstrate minimal antitumor activity in a population comprised predominantly of previously treated patients with metastatic RCC. Both individuals who responded had lung lesion(s) and a performance status of 51. These latter characteristics have been associated with response to other biologic response modifiers.‘‘ The hematopoietic effects of GM-CSFrhwere similar to those reported by previous investigators.”” The toxicity of E. coli-derived GM-CSFrhin this trial was mild, and repetitive administration was well tolerated. The majority of the Grade I11 and IV adverse events were disease related and were concentrated in a few individuals. One patient with liver involvement and a prior history of hypercalcemia developed increasing alkaline phosphatase (Grade 111) and recurrent hypercalcemia (Grade IV)while in the study. A second patient with progressive liver involvement developed Grade 111 elevation of alkaline phosphatase, serum glutamic pyruvic transaminase, and total bilirubin. An additional two patients developed transient neutropenia on day 1 of GM-CSF,,, therapy, which resolved by day 4 of the cycle. This was likely secondary to margination and has been described previously secondary to GM-CSF.23 Reports of tumor regression associated with GM-CSF administration have been uncommon. A Phase I study utilizing daily short infusions of GM-CSFrhnoted one objective response in a patient with metastatic sarco~na.’~ In a Phase I1 trial of concomitant subcutaneous IFN-a2b and GM-CSFrh,two partial responses in 15 patients were reported.” In a recent report by Edmonson et al,‘“ 15 patients with advanced and refractory sarcomas receiving chemotherapy (ifosfamide, mesna, doxorubicin, and cisplatin, withlwithout mitomycin C) received GM-CSF,,, post-therapy, and, in several individuals, it was also administered prior to treatment. Responses occurred in 10 of 15 (66%)patients, and the authors speculate that GMCSF,, may have played a role in producing this unexpectedly high response rate in this group of patients, who are generally refractory to chemotherapy. The antitumor effects of GM-CSF have been best demonstrated in murine tumor models in which mice were vaccinated with irradiated tumor cells transduced with the GM-CSF gene.” Paracrine production of this cytokine appeared to mediate protection from secondary challenge with wild type tumor cells and produced tumor 1152 CANCER March 15,1996 I Volume 77 / Number 6 mononuclear cell infiltrates consisting of b o t h CD4' and CD8' lymphocytes. These vaccinated animals developed specific and long-lasting antitumor immunity. In a subseq u e n t report," gelatin-chondroitin sulfate microspheres containing GM-CSF had effects similar t o genetically altered t u m o r cells. These investigators have speculated that the antitumor effects seen were related t o t h e enh a n c e m e n t of dendritic cell numbers and function by GM-CSF." Phase I trials in human malignancies, including renal cell carcinoma, are now underway utilizing genetically altered autologous t u m o r cells. In the present study, clear but short-lived antitumor effects were seen in several patients receiving GM-CSF,,,. It is possible that these responses were unrelated to GMCSFrh administration and were instances of s p o n t a n e o u s r e g r e s s i ~ n . *The ~ ~ ~rarity ~ of this phenomenon3' makes this unlikely, however. The mechanisms responsible for t h e t u m o r regressions observed in GM-CSF-treated patients are unclear. Preclinical studies suggest that CD8' and CD4' lymphocytes andlor monocytes'8 may be involved. In addition, t h e possibility that GM-CSF enhances t h e numbers andlor function of dendritic or antigen-processing cells is also a c o n ~ i d e r a t i o n . Systemic ~' administration of GM-CSF,h t o patients with large t u m o r burdens is clearly different than t h e murine model of paracrine GM-CSF production reported by Dranoff et al." It is of interest, however, that clinical t u m o r responses were seen in the present study and developed in patients with prior therapy. The responses were brief, b u t t h e toxicity of GMCSF,h, even in patients receiving repetitive cycles, was mild. The antitumor activity of this cytokine appears minimal, b u t additional trials of GM-CSF in previously untreated patients, as a part of combination regimens or administered with autologous t u m o r vaccines, may be avenues worthy of exploration. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1. Groopman JE, Molina JM, Scadden DT. Hematopoietic growth factors. Biology and clinical applications. N EnglJ Med 1989;32 1:1449-59. 2. Metcalf D,Begley CG, Johnson GR, Nicola NA, Vadas MA, Lopez AF, et al. Biologic properties in vitro of a recombinant human granulocyte-macrophage colony-stimulating factor. Blood 1986;67:37-45. 3. Hesketh PI, Sullivan R, Valeri CR. McCarroll LA.The production of granulocyte-monocyte colony-stimulating activity by isolated human T lymphocyte subpopulations. Blood 1984;63:1141-6. 4. Oster W, Lindemann A, Mertelsmann R, Herrmann F. Production of macrophage-, granulocyte-, granulocyte-macrophage-, and multi-colony-stimulating factor by peripheral blood cells. Eur 1lmmunol 1989; 19:543-7. 5. Zsebo KM,Yuschenkoff VN, Schiffer S, Chang D. McCall E, Dinarello CA,et al. Vascular endothelial cells and granulopoiesis: Interleukin-1 stimulates release of G-CSF and GMCSF. Blood 1988;71:99-103. 6. Wodnar-Filipowicz A, Heusser CH, Moroni C. Production of 16. 17. 18. 19. the hematopoietic growth factors GM-CSF and interleukin3 by mast cells in response to IgE receptor-mediated activation. Nature 1989;339:150-2. Zucali JR, Dinarello CA, Oblon DJ, Gross MA, Anderson L, Weiner RS. Interleukin 1 stimulates fibroblasts to produce granulocyte-macrophage colony-stimulating activity and prostaglandin E2. J Clin Invest 1986;77:1857-63. Cantrell MA,Anderson D, Cerretti DP, Price V, McKereghan K, Tushinski RJ, et al. Cloning, sequence, and expression of a human granulocyteImacrophage colony-stimulating factor. Proc Natl Acad Sci USA 1985;82:6250-4. Herrmann F, Schulz G, Lindemann A, Meyenburg W, Oster W, Krurnwieh D, et al. Hematopoietic responses in patients with advanced malignancy treated with recombinant human granulocyte-macrophage colony-stimulating factor. I Clin Oncol 1989;7:159-67. Lieschke GJ, Maher D, O'Connor M, Green M, Sheridan W, Rallings M, et al. Phase I study of intravenously administered bacterially synthesized granulocyte-macrophage colonystimulating factor and comparison with subcutaneous administration. Cancer Res 1990;50606- 14. Bukowski RM, Murthy S, McLain D, Finke J, Andresen S, Tubbs R, et al. Phase I trial of recombinant granulocytemacrophage colony-stimulating factor in patients with lung cancer: Clinical and immunologic effects. 1 Immunother 1993;13967-74. Sisson SD. Dinarello CA. Production of interleukin- 1 alpha, interleukin-1 beta and tumor necrosis factor by human mononuclear cells stimulated with granulocyte-macrophage colony-stimulating factor. Blood 1988;721368-74. Thomassen MI, Ahmad M, Bama BP, Antal J, Wiedemann HP, Meeker DP, Klein J, et al. Induction of cytokine messenger RNA and secretion in alveolar macrophages and blood monocytes from patients with lung cancer receiving granulocyte-macrophage colony-stimulating factor therapy. Cancer Res 1991;51:857-62. Reed SG, Nathan CF, Phil DL, Rodricks P, Shanebeck K, Conlon PI, et al. Recombinant granulocytelmacrophage colony-stimulating factor activates macrophages to inhibit Trypanosoma cruzi and release hydrogen peroxide. Comparison with interferon gamma. 1Exp Med 1987;166:173446. Grabstein KH, Urdal DL, Tushinski RJ, Mochizuki DY, Price VL,Cantrell MA, et al. Induction of macrophage tumoricidal activity by granulocyte-macrophage colony-stimulating factor. Science 1986;232:506-8. Foulke RS, Marshall MH, Trotta P, Von Hoff DD. In vitro assessment of the effects of granulocyte-macrophage colony stimulating factor on primary human tumors and derived lines. Cancer Res 1990;50:6264-7. Cannistra SA, Rarnbaldi A, Spriggs DR, Herrmann F, Kufe D, Griffin JD. Human granulocyte-macrophage colony-stimulating factor induces expression of the tumor necrosis factor gene by the U937 cell line and by normal monocytes. J Clin Invest 1987;79:1720-8. Dranoff G, Jaffee E, Lazenby A, Golumbek P, Levitsky H, Brose K, et al. Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colonystimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Nut1 Acud Sci USA 1993; 90:3539-43. Quesada JR, Rios A, Swanson D, Trown P, Gutterman JU. Antitumor activity of recombinant-derived interferon alpha in metastatic renal cell carcinoma. 1 Clin Oncol 1985;3: 1522-8. Phase II Trial GM-CSF in Metastatic Renal Cell CarcinomaMlos et al. 20. Bukowski RM, Goodman P, Crawford ED, Sergi JS, Redman BG, Whitehead RP. Phase 11 trial of high-dose intermittent 21. 22. 23. 24. 25. interleukin-2 in metastatic renal cell carcinoma: A Southwest Oncology Group study. J Natl Cancer Inst 1990;82:1434. Budd GT, Murthy S, Finke I, Sergi J, Gibson V, Medendorp S, et al. Phase I trial of high dose bolus interleukin-2 and interferon alpha-2a in patients with metastatic malignancy. / Cliri Oncol 1992;102304-9. de Forges A, Rey A. Klink M, Ghosn M, Kramar A, Droz JP. Prognostic factors of adult metastatic renal carcinoma: A multivariate analysis. Sernin Surg Oncol 1988;4:149-54. Deverew S, Bull HA, Campos-Costa D, Saib R, Linch DC. Granulocyte macrophage colony stimulating factor induced changes in cellular adhesion molecule expression and adhesion to endothelium: In-vitro and in-vivo studies in man. Br I Haernatol 1989;71:323-30. Steward WP, Scarffe JH, Austin R, Bonnem E, Thatcher N. Morgenstern G, et al. Recombinant human granulocyte macrophage colony stimulating factor (rhGM-CSF)given as daily short infusions-A Phase I dose-toxicity study. Br I Cancer 1989;59:142-5. O’Donnell RT, Dea G, Meyers FJ. A Phase 11 trial of concomitant interferon-alpha-2b and granulocyte macrophage colony-stimulating factor in patients with advanced renal cell cancer. J Irnmunother 1995; 1758-61. 1153 26. Edmonson JH, Long HJ, Kvols LK. Cytotoxic drugs plus sub- 27. 28. 29. 30. 31 cutaneous granulocyte-macrophage colony stimulating factor: Can molgramostim enhance antisarcoma therapy. J Nut1 Cancer Inst 1994;86:312. Golumbek PT,Azhari R, Jaffee EM, Levitsky HI, Lazenby A, Leong K, et al. Controlled release, biodegradable cytokine depots: A new approach in cancer vaccine design. Cancer Res 1993;53:5841-4. Huang AY, Golumbek P, Ahmadzadeh M, Jaffee E, Pardoll D, Levitsky H. Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens. Science 1994;264: 961 -5. Oliver RT, Miller RMM. Mehta A, Barnett MI. A Phase I1 study of surveillance in patients with metastatic renal cell carcinoma and assessment of response of such patients to therapy on progression. Mol Biofher 1988;1:14-20. Montie JE, Stewart BH, Straffon RA, Banowsky LH, Hewitt CB, Montague DK. The role of adjunctive nephrectomy in patients with metastatic renal cell carcinoma. J Urol 1977; 117:272-5. lnaba K, Inaba M, Romani N, Aya H, Deguchi M, Ikehara S, et al. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony stimulating factor. J Exp Med 1992;176:1693-702.