(CANCER RESEARCH 49, 6432-6436, November 15, 1989] Phase I Clinical Trial of Interleukin 2 and a-Interferon: Toxicity and Immunologie Effects1
G. T. Budd,2 B. Osgood, B. Barna, J. M. Boyett, J. Finke, S. V. Medendorp, S. Murthy, C. Novak, J. Sergi, R. Tubbs, and R. M. Bukowski The Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio 44195
ABSTRACT Interleukin-2 acts to maintain T-lymphocyte growth, aug ment NK activity, induce T-helper cell function, and activate Because recombinant interleukin 2 (rIL-2) and recombinant a-inter- antigen-specific cytotoxic T-lymphocytes (13-15). Further feron (rIFN-a) exhibit synergistic antitumor activity in C3HMT1820 more, IL-2 can induce nonspecific cytotoxicity against NK- I -coll lymphoma and B16 melanoma tumor systems, we have performed resistant as well as NIC-sensitive tumor targets, an activity a Phase I study of this combination in 55 patients with advanced malig nancies for whom no standard therapy exists. Successive groups of >4 attributed to a cell population referred to as LAK cells (1). patients have been entered into 12 dose levels (1A-3D), with dose levels Investigations of this phenomenon have demonstrated that this 1-3 referring to doses of rIL-2 of 0.1, 0.5, and 2.0 x 10' iinits/nr. LAK activity is largely mediated by IL-2-activated NK cells respectively, and dose levels A-D referring to doses of recombinant (16-19). human «2a-interferon(rHu!FN-a2a) of 0, 0.1, 1.0, and 10.0 x 10" units/ Because both IFN-a and IL-2 can induce cytotoxicity me ur. Both agents were given on Mondays, Wednesdays, and Fridays, with diated by NK cells, investigations of combination therapy with rIL-2 being given as i.v. bolus injections and rHuIFN-a2a being given these two cytokines are of interest. In one such study, combi intramuscularly. Myelosuppression was dose-limiting and was related nation therapy with IL-2 and IFN-a A/D resulted in synergistic primarily to the dose of ri lui !•V«2;i.The maximum-tolerated dose level was reached at a dose of rIL-2 of 2.0 x I(I'' units/nr and of ri lui I N-«2a augmentation of NK cell activity in a murine system (20). of 10.0 x ID" units/in' (dose level 3D). At this dose level, 3/6 patients Studies of the in vivo antitumor effect of the same combination developed grade 3 neutropenia (absolute granulocyte count
Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research. PHASE I TRIAL OF rIL-2 AND rHuIFNo2a nervous system metastasis or seizure disorder. In addition, all patients Ficoll-Hypaque separation of heparinized blood were evaluated in 4-h were required to have laboratory studies with values within the following 51Cr-release assays. NK activity was determined against "Cr-labeled ranges no more than 14 days prior to the initiation of treatment: WBC NK-sensitive K562 erythroleukemia cells (American Type Culture Col z3 x lO'/Iiter; granulocytes >1 x 109/Hter; platelets >100 x lO'/Hter; lection) as previously described (21). LAK activity was evaluated using hemoglobin >9.5 gm/100 ml; prothrombin time <14.0 s; activated "Cr-labeled NK-insensitive Daudi lymphoma cells (American Type partial thrombin time <31.0 s; creatinine si.5 mg/dl; bilirubin (total) Culture Collection) (22). Assays were carried out in microtiter trays <1.5 mg/dl; carbon monoxide diffusing capacity >50% of predicted; using 10,000 target cells per well and effector/target ratios of 3:1 to and a 1-s forced expiratory volume/forced vital capacity >80% pre 50:1. Percentage specific lysis ("activity") at each ratio was determined dicted. Informed consent was obtained in accordance with institutional after harvesting assays with a Skatron device (Skatron, Sterling, VA). and NCI guidelines. Results were expressed as lytic units per 1 x IO7 MNL with one lytic Treatment and Study Design. The treatment schema is outlined in unit defined as the number of effector cells producing 30% specific lysis Table 1. The rHu!FN-«2a was reconstituted with sterile water and of target cells in 4 h ("activity"). Results were also analyzed at the administered by i.m. injection; rHuIL-2 was reconstituted with sterile effector/target ratio of 25:1, expressed as a percentage cytotoxicity water and administered by i.v. bolus injection over 5 min immediately ("cytotoxicity"). following rHu!FN-«2a. Both drugs were administered three times Statistical Methods. Multiple readings of immune parameters were weekly on Mondays, Wednesdays, and Fridays. Treatments were con taken on each study patient during the pretreatment interval and the tinued weekly for a minimum of 4 weeks. Prior to each treatment treatment period. The two pretreatment levels were averaged and the patients were given a single oral dose of acetaminophen (650 mg); percentage change from baseline calculated for each parameter for each meperidine 50 mg was given i.v. in the event of severe rigors. In the recording. For LAK activity, absolute changes were calculated rather presence of stable disease or objective tumor regression, therapy was than percentage changes, as the baseline values were often zero. The continued until there was evidence of disease progression or unaccept Kruskal-Wallis test was used to test for evidence of an IFN dose effect able toxicity. Four to six patients were entered at each dose level and in mean percentage change from baseline within each dose level of IL- observed for toxicity before additional patients were entered at the next 2 (23). In addition, the Kruskal-Wallis test was used to test for an higher dose level. Response and toxicity were graded using ECOG overall IFN effect and an overall IL-2 effect. criteria, and the maximum tolerated dose (MTD) was defined as that level at which three or more of six patients experienced Grade III or IV toxicity (excluding fever). Grade III hypotension was defined as a RESULTS blood pressure of <70/50 mm Hg or a drop in the diastolic blood pressure of 30-39 mm Hg as compared to baseline and Grade IV Patient Characteristics. A total of 55 patients were entered hypotension as a blood pressure of -=60/40 mm Hg or a decrease in the on study, 40 males and 15 females with a mean age of 55 years diastolic blood pressures of >40 mm Hg. (range, 30-77). Twenty-three patients had a performance status Study Monitoring. Within 7 days of the start of therapy, all patients of 0 and 32 had a performance status of 1. Eighteen patients had a history and physical examination, complete blood count, serum had received no previous nonsurgical therapy for their malig chemistry profile, urinalysis, electrocardiogram, pulmonary function, nancies, 19 had received chemotherapy alone, seven had re and radiological studies to evaluate their disease. Follow-up hemato- ceived radiotherapy alone, and 11 had received both chemo logical studies were performed prior to each drug administration, with therapy and radiotherapy. The numbers of patients treated with weekly biochemical monitoring and monthly physical and radiological examinations. Immunological studies of peripheral blood were per particular tumor types were as follows: colorectal cancer (14), renal cell carcinoma (12), non-small cell lung cancer (8), non- formed twice prior to therapy and on days 5, 15, and 28 of treatment. On days when both immunological studies and therapy were called for, colorectal gastrointestinal tumors (7), malignant melanoma (6), immunologie-ai studies were drawn prior to the administration of ther unknown primary site (3), sarcoma (2), breast cancer (1), pros- tatic cancer (1), and Kaposi's Sarcoma in conjunction with the apy. These studies included determinations of peripheral blood lym phocyte subsets, NK activity and LAK activity. Most patients were acquired immunodeficiency syndrome (1). The number of pa treated in the outpatient department, where their vital signs and phys tients entered on each dose level ranged from four to six (Table ical condition were closely observed during treatment and for 6 h 1). thereafter. Patients were evaluated for the development of toxicity 48 Toxicity. All patients suffered fever, chills, and malaise. Three h after receiving each treatment and were evaluated every 4 weeks for patients suffered fevers of greater than 40°C,all at doses of tumor response. Treatment was continued until disease progression in rHu!FN-a2a of 10 x IO6units/M2 (one each at dose levels ID, the absence of intolerable toxicity. Immunological Studies. Forty-eight of the 55 patients donated serial 2D, and 3D). Nausea was mild to moderate in severity, with samples of peripheral blood for immunological studies. The methods occasional vomiting being noted. Occasionally patients devel used in immunological monitoring are described below. oped stomatitis or diarrhea, which was at worst Grade 2. Grade Quantitations of Lymphocyte Subpopulations. Mononuclear leuko 3 elevations of the serum alkaline phosphatase or bilirubin were cytes (MNL) were separated from heparinized blood by standard tech seen in eight instances, all in patients who had known liver niques involving centrifugation over Ficoll-Hypaque. After phagocytic métastases.Mild hypotension was produced in 10 patients. A cells were removed with carbonyl iron, aliquots of lymphocytes were single patient was treated with a vasopressor; hypotension in stained simultaneously with monoclonal antibodies conjugated to flu- all other cases responded rapidly to fluid replacement. orescein isothiocyanate and phycoerythrin. Controls were stained with All patients experienced generalized fatigue in conjunction similarly labeled irrelevant IgG of the corresponding isotypes. Samples were analyzed by flow cytometry using a FACS 440 (Becton-Dickenson, with this treatment. The median decrement in SWOG Perform ance Status was 0.4 points (range, 0-3; standard deviation, Mountain View, CA) fitted with a logorithmic amplifier. The total white blood cell count, determined from a concurrent complete blood Table 1 Maximum grade neutropenia per dose level; number of patients with count, was used to calculate the absolute numbers of subset cells per toxicity Grades 1-4 microliter of blood. Monoclonal antibodies directed against the follow (Number Grade 3-4)/number of patients entered at that dose level. ing markers were used: B-Cells (Leu 12+); T-cells (Leu 4+); helper T- units/M2)00/4of rHuIFN-a2a (IO6 cells (Leu 3+/S-); inducer T-cells (Leu 3+/8+); suppressor T-cells DoseofIL-2(10* (Leu 2+/15+); cytolytic T-cells (Leu 2+/15-); activated T-cells (Leu BRMP units/M2)0.10.5 4+/HLA-DR+, Leu 4+/IL-2 R+); and NK (Leu 11+). Evaluation of Cell-mediated Cytolytic Activities (NK Activity and LAK 2/40/40.10/41/5 1/42/510.02/42(1)/6 Activity). Cytolytic activities of mononuclear lymphocytes derived from 2.0Dose 0/51.00/4 4(3)/6 6433
Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research. PHASE I TRIAL OF rIL-2 AND rHuIFNa2a
0.74), with five patients experiencing a Performance Status Statistically significant relationships between NK activity and decrease of two categories and one patient a PS decrease of 1 both cytokines were found in exploratory analysis, as presented point. Patients receiving this treatment tended to lose weight in Table 3. The relationship between IL-2 dose, rHu!FN-<*2a over time, with the mean weight loss being 3.6 kg (range, 0- dose and NK activity is depicted in Fig. 1. To generate this 12.9 kg; standard deviation, 2.9 kg). These changes could not figure, the percentage difference between the mean of pretreat be demonstrated to be statistically significantly related to cy- ment assays and the mean of all on-treatment assays was tokine dose, but three of the six patients with a PS decrement calculated for each patient. The median of these mean percent of two or more points were treated with rIL-2 at the highest age changes for the patients at each dose level were graphed as dose of 2.0 x IO6 units/M2. At the doses studied, these effects a function of the doses of the two cytokines. NK activity tended were not dose-limiting, and repeated outpatient treatments were to increase with the dose of IL-2. At all dose levels of IL-2, possible. however, the dose of rHu!FN-t*2a associated with maximal NK Myelosuppression proved dose-limiting, with neutropenia activity was 0.1 x IO6 units/M2. Changes in lymphokine-acti- and thrombopenia being related primarily to the dose of vated killer activity were also observed, but these changes could rHuIFN-a2a (Table 1). The maximum tolerated dose level was not be demonstrated to be statistically significantly related to reached at a dose of IL-2 of 2.0 x IO6 units/M2 and rHuIFN- the dose of either cytokine (Table 3, Fig. 2). «2aof 10 x IO6 units/M2. At this dose level, three of six Because of the small sample size at any given dose-level and patients developed Grade 3 neutropenia (absolute granulocyte the inherent variability in the immunological parameters being count less than 1 x 109/liter. No patient developed evidence of monitored, considerable variability in the immunological effects infection and no bleeding episodes were noted. Myelosuppres was observed at each dose level. To illustrate this, the standard sion was transient, reversing in 48-72 h, and was generally errors (of the mean) for NK activity are displayed in Fig. 1 and observed at the end of each weekly cycle of treatment. related to the medians of the mean percentage change. Varia Antitumor Activity. Three partial responses and a single mi bility of a similar magnitude was present in all of the immu nor response were observed (Table 2). No consistent relation nological parameters studied, and, as discussed below, any ship was present between anti-tumor effect and dose-level or conclusions must be regarded as preliminary. immunologie changes. Immunological Effects. Table 3 summarizes the results of the Kruskal-Wallis Tests relating the listed immune parameters DISCUSSION and the dose of each cytokine. Statistically significant (P < The principal aim of this study was to examine the toxicity 0.05) changes as a function of dose were seen in white blood of the combination of rHuIL-2 and rHuIFN-a2a. We have count for both rHuIFN-«2a and IL-2 and in total lymphocyte found that these two agents may be given together safely, with count as a function of IL-2 dose. The changes in the granulocyte count as a function of the doses of the two cytokines are summarized in Table 1, and these changes constituted the dose- limiting toxicity defining the maximum tolerated dose. Statis tically significant changes were seen in the following lympho cyte subsets as a function of IL-2 dose: Leu 4+ (total T-cells), Leu 3+/8~ (T-helpers), and Leu 2+/15~ (cytotoxic T-lympho- cytes). These changes parallelled the changes seen in the total lymphocyte count, which, at the highest dose levels of IL-2, increased 37% on the days that lymphocyte subsets were deter mined, but which did not change significantly when the lym phocyte counts on all days that complete blood counts were drawn were considered. "Rebound" lymphocytosis was ob served neither upon the completion of therapy nor 72 h after the completion of each weekly treatment. 10.0
Table 2 Summary of antitumor responses Fig. 1. Median of the mean percentage change in NK activity with therapy for patients at each dose level of rIL-2 and rHuIFN-o2a (see text for further level(IL-2/IFN-a2a)IB explanation). TumortypeMelanoma ofdiseaseLymph node 28.4 6 weeks (0.1/1.0) cm2—»1.8cmJ Melanoma Soft tissue 4.8 cm2 —¿ Minor 10 weeks 20(0.5/10.0) 3.3 cm2 Renal cell Lymph node 19.2 Partial 12 weeks 3C (2.0/1.0) cm2 -> 9.0 cm2 BreastSites Liver 7.0 cm2 -> 3.5 Partial 4 weeksDose30(2.0/10.0) cm2ResponsePartial
Table 3 Summary of Kruskal-Wallis tests ValuesParameterWBCNK P
activity 0.03 0.18 NK cytotoxicity 0.10 0.02 Fig. 2. Median of the mean percentage change in LAK activity with therapy LAK activity (difference) 0.81 0.23 for patients at each dose level of rIL-2 and rHu!FN-«2a (see text for further LAK cytotoxicityIL-20.86 0.70IFN0.0001 0.46 explanation.) 6434
Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research. PHASE l TRIAL OF rIL-2 AND rHiiIFNo2a myelosuppression proving dose-limiting at a dose of IL-2 of 2.0 dose-limiting toxicity of the combination of IL-2 and rHuIFN- X 10" units/M2 and rHuIFN-«2a of 10 x IO6 units/M2. A «2a,when given according to a thrice weekly bolus schedule. statistically significant relationship exists between the dose of When given together, the maximum tolerated doses of IL-2 and rHu!FN-«2a and the degree of neutropenia. Further dose es rHu!FN-«2a were 2.0 x IO6units/M2 i.v. and 10.0 x IO6units/ calations of IL-2 may well be possible at doses of rHuIFN-a2a M2 i.m., respectively. Higher doses may be possible by ignoring of less than 10 x IO6 units/M2. Myelosuppression proved asymptomatic myelosuppression. NK cell activity seemed pos rapidly reversible and was not associated with significant se itively related to the dose of IL-2 and maximal at an rHuIFN- quelae in this study. Given the expertise in dealing with mye ala dose of 0.1 x IO6 units/M2, suggesting that, among those losuppression which has resulted from a long clinical experience studied, the "optimal biological dose" with respect to NK with myelosuppressive chemotherapeutic agents, further dose activity is the combination of IL-2 2.0 x IO6 units/M2 and escalations may be possible despite significant myelosuppres rHuIFN-«2a0.1 x IO6units/M2. These conclusions result from sion. Additionally, alternative schedules of administration of an exploratory analysis that must be viewed as hypothesis- these two agents may alter their toxicity. At higher doses of generating rather than as conclusive. Our findings do justify these agents, generalized fatigue, weight loss, and a decrease in further studies designed to prospectively compare higher and Performance Status may prove dose-limiting, as has been de lower doses of rHu!FN-«2a and IL-2 in order to more defini scribed in a Phase I trial of rIL-2 and IFN-/3 (24). Currently, tively determine the relationships between (a) cytokine dose we are exploring both of these approaches in two Phase I trials and modulation of given immunological parameters, (b) biolog based upon the results reported here. One trial involves further ical response modulation and antitumor effect, and (c) dose and dose escalation of IL-2, using the same schedule of administra antitumor effect. tion as described above, while the second trial involves IL-2 administered by continuous i.v. infusion. ACKNOWLEDGMENTS While this trial was not designed to explore the efficacy of this combination, some evidence of antitumor activity was We would like to acknowledge the help of R. Valenzuela in the produced, with three partial and one minor response being performance of the lymphocyte subset analysis, B. Jacobs in the per observed in 55 patients. One partial and one minor response formance of NK and LAK activity assays, and B. Weiss in data were produced in six patients with malignant melanoma, sug management. gesting that this tumor type should be explored in subsequent trials of this combination. Partial responses were also seen in REFERENCES patients with renal cell and breast cancers; Phase II trials in these histologies should also be considered, particularly in renal 1. Grimm, E. A., Mazumder, A., Zhang, H. Z., and Rosenberg, S. A. Lympho- cell carcinoma where both IL-2 and rHuIFN-a2a have minor kine-activated killer cell phenomenon. Lysis of natural killer resistant fresh solid tumor cells by interleukin 2 activated autologous human peripheral single-agent activity. Whether the combination of rHuIL-2 and blood lymphocytes. J. Exp. Med., 755: 1823-1841, 1982. rHuIFN-a2a is superior to single agent cytokine therapy can 2. Herberman, R. B., Ortaldo, J. R., Mantovani, A., Hobbs, D. S., Kung, H.- F., and Pestka, S. Effect of human recombinant interferon on cytotoxic be determined only by Phase II and III studies. activity of natural killer (NK) cells and monocytes. Cell Immunol., 67: 160- Exploratory analysis of relevant immunological parameters 167, 1982. suggested the following statistically significant relationships: 3. Brunda. M. J., and Rosenbaum, D. Modulation of murine natural killer cell activity in vitro and in vivo by recombinant human interferons. Cancer Res., (a) neutropenia was related to the dose of rHuIFN-a2a, with «.•597-601,1984. higher doses of IFN being associated with more severe neutro 4. Brunda, M. J., and Rosenbaum, D., and Stern, L. Inhibition of experimen penia; (b) natural killer activity was related to the dose of IL-2, tally-induced murine métastasesbyrecombinant alpha interferon: correlation between the modulatory effect of interferon treatment on natural killer cell with higher doses of IL-2 being associated with greater degrees activity and inhibition of métastases.Int. J. Cancer, 24:421 -426, 1984. of augmentation of NK activity; (c) natural killer activity was 5. Goldstein. D., and Laszlo. J. The role of interferon in cancer therapy: a also related to the dose of rHuIFN-a2a, but the greatest degree current perspective. CA-A J. Clinicians, 38(5): 258-277, 1988. 6. Goldstein, D., and Laszlo, J. Interferon therapy in cancer: from imaginon to of augmentation was seen at the lowest dose of rHuIFN-«2a, interferon. Cancer Res, 46:4315-4329, 1986. 0.1 x IO6units/M2. While consistent with known relationships, 7. Suzuki, R., Simuli. S., Ebina, N., and Kumagai, K. Suppression of alloim- mune cytotoxic T lymphocyte (CTL) generation by depletion of NK cells and the above results must be interpreted with caution, given the restoration by interferon and/or interleukin 2. J. Immunol., 134:2139-2148, risks inherent in multiple subset analyses. 1985. The finding of an inverse relationship between rHuIFN-a2a 8. Herberman, R. B. Natural Cell Mediated Immunity Against Tumors. New York: Academic Press, 1980. dose and NK activity may seem anomalous, given the reports 9. Lindahl. P., Leary, P., and Gresser, I. Enhancement by interferon of the of increasing augmentation of NK activity with increasing doses specific cytotoxicity of sensitized lymphocytes. Proc. Nati. Acad. Sci. USA, of both IL-2 and IFN-a A/D in vitro (25). However, in vivo 69:721-725, 1972. 10. Murray, J. L., Pillow, J. K., and Rosenblum, M. G., et al. Differential in studies in cancer patients have shown an inverse relationship vitro effects of alpha recombinant interferon (rIFN-alpha-A) and gamma between the dose of a-interferon and NK activation, despite a recombinant interferon (rIFN-gamma) on the expression of melanoma-as sociated antigens (MAA) P97 and 240 Kd on melanoma cell line Hs294t. positive correlation between NK cell activation and interferon Proc. Am. Assoc. Cancer Res., 27: 313, 1986. concentration in in vitro systems. Lower doses of IFN-a were 11. Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research. PHASE I TRIAL OF rIL-2 AND rHuIFNo2a the growth and differentiation of lectin-stimulated cytolytic T lymphocyte American Society for Microbiology, 1986. precursors. J. Immunol., 134: 1644-1651, 1985. 22. Grimm, E. A., Muzumder, A., Zhang, H. Z., and Rosenber, S. A. Lympho 16. Phillips, J. II.. and Lanier, L. I . Dissection of the lympokine-activated killer kine-activated killer cell phenomenon: lysis of natural killer-resistant fresh phenomenon. Relative contribution of peripheral blood natural killer cells solid tumor cells by Interleukin-2-activated autologous human peripheral and T-lymphocytes to cytolysis. J. Exp. Med., 164: 814-825, 1986. blood lymphocytes. J. Exp. Med., 755: 1823-1841, 1982. 17. Ortaldo, J. F., Mason, A., and Overton, R. Lymphokine-activated killer cells. 23. Kruskal, W. H., and Wallis, W. A. Use of ranks in one-criterion variance Analysis of progenitors and effectors. J. Exp. Med., 164: 1193-1205, 1986. analysis. J. Am. Stat. Assoc., 47: 583-621, 1952. 18. Itoh, K., Tilden, A. B., and Balch, C. M. Lysis of human solid tumor cells 24. Krigel, R. L., Padavic-Shaller, K. A., Rudolph, A. R., Litwin, S., Konrad, by lymphokine-activated natural killer cells. J. Immunol., 136: 3910-3915, M., Bradley, E. C., and Comis, R. L. A Phase 1 study of recombinant 1986. interleukin plus recombinant ff-interferon. Cancer Res., 48: 3875-3881, 19. Phillips, J. H., Gemlo, B. T., Myers, W. W., Rayner, A. A., and Lanier, L. 1988. L. In vivo and in vitro activation of natural killer cells in advanced cancer 25. Brunda, M. J., Tarnowski, D., and Davateks, V. Interaction of recombinant patients undergoing combined recombinant interleukin-2 and LAK cell ther interferons with recombinant interleukin-2: differential effects on natural apy. J. Clin. Oncol., 5: 1933-1941, 1987. killer cell activity and interleukin-2 activated killer cells. Int. J. Cancer, 37: 20. Brunda, M. J., Tarnowski, D., and Davateks, V. Interaction of recombinant 787-793, 1986. interferons with recombinant interleukin-2: differential effects on natural 26. Edwards, B. S., Merritt, J. A., Fuhlbrigge, R. C., and Borden. E. C. Low killer cell activity and interleukin-2 activated killer cells. Int. J. Cancer, 37: doses of Interferon Alpha result in more effective clinical natural killer cell 787-793, 1986. activation. J. Clin. Invest., 75: 1908-1913, 1985. 21. Herberman, R. B. Natural killer cell activity and antibody-dependent cell- 27. Edwards, B. S., Hawkins, M. J., and Borden, E. C. Comparative in vivo and mediated cytotoxicity. In: N. R. Rose, H. Friedman, and J. L. Fahey (eds.). in vitro activation of human natural killer cells by two recombinant alpha- Manual of Clinical Immunology, Ed. 3, pp. 308-3114. Washington, DC, interferons differing in antiviral activity. Cancer Res., 44: 3135-3139, 1984. 6436 Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research. Phase I Clinical Trial of Interleukin 2 and α-Interferon: Toxicity and Immunologic Effects G. T. Budd, B. Osgood, B. Barna, et al. Cancer Res 1989;49:6432-6436. Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/49/22/6432 E-mail alerts Sign up to receive free email-alerts related to this article or journal. Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected]. Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/49/22/6432. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site. Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1989 American Association for Cancer Research.