- Email: [email protected]
Biochemotherapy with thymosin α1, interleukin-2 and dacarbazine in patients with metastatic melanoma: Clinical and immunological effects
Annals of Oncology 5:741-746, 1994. © 1994 Kluwer Academic Publishers. Printed in the Netherlands.
Original article Biochemotherapy with thymosin a l , interleukin-2 and dacarbazine in patients with metastatic melanoma: Clinical and immunological effects M. Lopez, S. Carpano, R. Cavaliere, L. Di Lauro, F. Ameglio,1 G. Vitelli, A. M. Frasca, P. Vici, F. Pignatti, M. Rosselli, G. Rasi2 & E. Garaci2 Regina Elena Institute for Cancer Research, Rome; '5. Gallicano Institute, Rome; 2'Tor Vergala' University, Rome, Italy
months and median survival was 11 months. Side effects were predominantly caused by IL-2. Treatment was tolerated Background: DTIC and interleukin-2 (IL-2), as single agents, reasonably well, and there was no overlapping toxicity or have a limited anti-tumor activity in patients with metastatic interference between chemotherapy and biotherapy. Baseline melanoma. Experimentally, thymosin a l (TA1) may modu- sCD4 levels seem to correlate to tumor burden. Patients late the action of IL-2. We investigated the clinical and im- benefiting from treatment had lower sCD4 and higher sCD8 munological effects of a combination with these three agents. than did progressing patients. Patients and methods: Forty-six patients with measurable Conclusions: The combination of DTIC + TA1 + IL-2 is metastatic melanoma were treated with DTIC 850 mg IV on active in the treatment of advanced melanoma, with acceptday 1, TA1 2 mg s.c. on days 4 to 7, and IL-2 18 MU/m2/d able toxicity. However, even more active regimens are needby continuous intravenous infusion on days 8 to 12. Cycles ed, and the interactions between thymic hormones and cytowere repeated every 3 weeks. kines should be further explored. Results: Objective responses were obtained in 15 (36%) of 42 evaluable patients (CI at 95%: 22%-50%). Two patients experienced complete responses, and stable disease was ob- Key words: advanced melanoma, biochemotherapy, DTIC, served in five. The median time to progression was 5.5 interleukin-2, thymosin a l Summary
Introduction
sin a l (TA1) may induce a wide variety of biological effects, including stimulation of NK activity, proliferative response of CD4+ cells to various antigens, production of several cytokines (IL-1, IL-2, IFN, TNFa), and expression of IL-2 receptors [4]. All these findings suggest that this agent may play a role in the modulation of the immune response against cancer. Interestingly, a recent experimental study by Mastino et al. reported a significantly improved survival of mice bearing Lewis lung carcinoma treated with combined cyclophosphamide (CTX), TA1 and IL-2 over that of mice receiving CTX, TA1 or IL-2 alone, or CTX plus a single immunomodulator [5]. Based on this clinical and experimental evidence, a phase II study was undertaken in patients with metastatic melanoma to evaluate the effectiveness and toxicity of dacarbazine combined with TA1 and CTV IL-2.
Metastatic melanoma is usually resistant to many forms of therapy. Only a small number of chemotherapeutic agents have demonstrated antitumor activity against this disease, and dacarbazine (DTIC) remains the most active single agent, with response rates of around 20%, usually of short duration and without impact on patient survival [1]. Combination chemotherapy has not proven to be clearly superior to DTIC alone [1]. In recent years, the availability of a number of cytokines along with a better understanding of the immune system has provided the opportunity to explore new approaches of treatment for melanoma. The most extensive experience has been gained with interferons (IFNj) and interleukin-2 (IL-2). When given alone, IL-2 was associated with response rates of up to 24%, which is similar to the rates observed when it was given in combination with LAK cells [2]. The regimen used in these trials included high-dose IL-2 by bolus injection, Patients and methods and was potentially associated with severe toxicity requiring intensive care unit (ICU) support for many pa- Patients tients. In an attempt to reduce toxicity and improve eligibility, patients were required to have a histologically or cytotherapeutic results, continuous intravenous (CIV) infu- For logically confirmed diagnosis of metastatic melanoma; bidimensionsion schedules have been developed [3]. ally measurable disease; age <75 years; World Health Organization There is an increasing body of evidence that thymo- (WHO) performance status (PS) <2; adequate bone marrow
742 (neutrophils > 1,500/uL; platelets > 100,000/(iL), renal (creatinine level < 1.5 mg/dL), and hepatic (bilirubin level < 1.5 mg/dL) function; no history or current evidence of cardiovascular disease; no infections requiring antibiotic therapy; no corticosteroids; and no brain metastases. Patients previously treated with chemotherapy, radiation therapy or interferon were not excluded provided that a recurrence had been documented and at least 4 weeks had elapsed since the end of the treatment. Forty-six patients were entered into the study and all were evaluable for toxicity. Four patients were considered not evaluable for response (refusal, 2; lost to follow-up, 1; incorrect diagnosis, 1). Patient characteristics are listed in Table 1. Twenty-five patients (54%) had >2 metastatic sites and 26 (56%) had visceral disease including lung in 22 patients and liver in 6 patients. Five patients had received prior chemotherapy, and six were given IFN in an adjuvant setting. All patients provided informed consent before admission to the trial.
1 week. If hematologic recovery was not complete by day 28, the DTIC dose was reduced by 50% for leukocyte counts between 2000/(iL and 2999/uL or platelets between 50,000/uL and 74,999/u.L. In the event of lower values, treatment was discontinued to allow recovery. Infusion of IL-2 was temporarily interrupted in the event of persistent severe hypotension, severe agitation or confusion, increase in serum bilirubin levels > 5 mg/dL, increase in serum creatinine levels >4,5 mg/dL, dyspnea at rest not reversed by binasal oxygen, bacterial sepsis, or other severe side effects. IL-2 doses were reduced by 50% if any of the following events had occurred in the previous cycle: hypotension that did not reverse within 48 h of stopping the agent, increase in serum creatinine levels >6 mg/dL, increase in serum bilirubin levels > 5 mg/dL or grade HI neurotoxicity. Treatment was definitively interrupted in instances of myocardial ischemia, grade IV neurotoxicity, increase in serum creatinine or bilirubin levels that failed to return to baseline values, or severe hypotension at reduced doses.
Treatment DTIC 850 mg/m2 IV over 30 minutes was given on day 1 with appropriate antiemetic therapy. Thymosin a l was kindly provided by Sclavo (Siena, Italy) in lyophilized vials containing 2 mg/vial, and was administered subcutaneously at a dose of 2 mg on days 4 to 7. IL-2 (Proleukin; Eurocetus Corporation, Amsterdam, The Netherlands) was given at a dose of 18 x 106 IU/mVd by a 24 h CTV on days 8 to 12. This required a central venous catheter, and the use of prophylactic antibiotics. Cycles were repeated every 3 weeks up to a maximum of 6 cycles. Treatment was discontinued in instances of disease progression, unacceptable toxicity or patient refusal. After completion of the treatment plan patients were followed until disease progression.
Dose modification No DTIC dose reduction was planned in instances of leukopenia (3000-3999 cells/uL) or thrombocytopenia (75,000-99,999 cells/ u.L) on days 21. Below these ranges, treatment was delayed by Table 1. Patient characteristics. Characteristics
No. of patients
Entered/evaluable Site of primary tumor Cutaneous Ocular Unknown Sex Male/Female Age (years) Median (range) Stage
46/42
in/iv Performance status (WHO) Median (range) Sites of disease Skin, soft tissues Lymph nodes Liver Lung Bone Adrenal Abdominal mass Spleen No. of metastatic sites 1/2/3/4
44 1 1 27/19 49.5 (22-74) 4/42
Supportive measures Paracetamol (650 mg x 6) or indomethacin (25 mg x 4-6) was used prophylactically to avoid fever, and ondansetron (8 mg x 3) was given to control nausea and vomiting. Hypotension was treated by intravenous colloid or crystalloid and/or dopamine (5-10 ng/kg/ min) as necessary.
Immunological studies Treatment-related immunological effects were evaluated in 22 patients (10 responders and 12 non-responders). Blood samples were taken before DTIC (day 1), TA1 (day 4) and IL-2 (day 8) administration, at the end of IL-2 infusion (day 12), and on the third day after completion of the treatment cycle (day 15). Lymphocyte phenotype (CD3, CD4, CD8, DR+, CD16, CD25, CD 19) was tested on fresh samples by FACS analysis, using commercially available monoclonal antibodies (Becton Dickinson, Mountain View, CA, USA). Serum aliquots were stored at -40 *C to permit all assays to be run in parallel simultaneously. Serum immunoglobulins (IgG, IgA, IgM) were determined by rate nephelometry (Beckman, Brea, CA, USA). IL-lp", IL-2, IL-6, IFN-gamma, and TNFa were measured by enzyme-amplified sensitivity immunoassays (Medgenix Diagnostics, Fleurus, Belgium), IL-4 was determined by sandwich enzyme immunoassay (Genzyme, Cambridge, MA, USA), beta-2-microglobulin (B2M) was determined by microparticle enzyme immunoassay (Abbott, Chicago, IL, USA), and soluble membrane molecules (sIL2R, sCD4, sCD8) were detected by sandwich enzyme immunoassays (T Cell Sciences, Cambridge, MA, USA). Data not normally distributed were assessed using the Anova and the Kruskal-Wallis tests. Statistical analysis was performed using the students t-test In the case of paired data a paired t-test was employed.
1(0-2) 19 20 6 22 3 2 4 1 21/20/1/4
Non-evaluable patients: refusal, 2; lost to follow-up, 1; incorrect diagnosis, 1.
Evaluations Before treatment, all patients underwent complete physical examination, chest x-ray, abdominal ultrasound or computer tomography (CT), and CT scans of the brain and chest. Hematologic and biochemical parameters were determined. The extent of disease was clinically assessed every 2 weeks. Lesions that were assessable by x-rays, ultrasonography or CT were checked every 2 cycles. Response and toxicity were evaluated according to WHO criteria (6]. For toxicities not reported in this system, the National Cancer Institute common toxicity criteria were used.
743 Response duration and survival were determined by the KaplanMeier method, and compared using the log-rank test. The Fisher's exact test was used to analyze differences in response rates and toxicity among subgroups of patients.
100
AJI pttianti: 5.6 months NC: 11.B monthi CR • PR: 4.7 months
Results 20 "
Response Objective responses were seen in 15 of 42 (36%; 95% confidence interval, 22% to 50%) patients. There were two complete responses (CR), in patients with skin and lung metastases, and 13 partial responses (PR). Five patients (12%) had stable disease (SD). Responses were seen in the following sites: lymph nodes, 3 of 20 (15%); skin and soft tissue, 7 of 19 (37%); lung, 10 of 22 (45%); liver, 1 of 6 (17%); abdominal masses, 1 of 4 (25%); adrenals, 1 of 2 (50%). No responses were seen in 3 patients with bone disease. Tumor regression was apparent after the first treatment cycle in 2 patients, and after the second and the third cycle in 3 and 10 patients, respectively. No correlation could be identified between patient pretreatment characteristics (previous treatment, performance status, visceral/non-visceral disease, number of metastatic sites, presence/absence of vitiligo) and likelihood of response to the biochemotherapy. After a maximum follow-up period of 29 months, the median time to progression (Fig. 1) for stabilized and responding patients is 5.5 months, for responding patients, 4.7 months, and 11.5 months for stable patients (CR + PR vs. SD: P = 0.03). The median survival is 11 months (Fig. 2), with 48% of patients surviving > 1 year. There is no statistical difference between responding patients and those with stable disease (P — 0.30). At present 9 patients are ah've. Toxicity All 46 patients were evaluable for toxicity, as shown in Table 2. Fatigue, fever, chills, skin rash, nausea and Table 2. Toxicity in 46 patients. Toxic effect
Fatigue Fever-chills Erythema-rash N&V Stomatitis Diarrhea Hypotension Anemia Thrombocytopenia Nephrotoxicity Hepatic Arrythmia Sepsis
WHO grade 0
1
2
2 3 23 21 32 27 2 41 40 36 44
10 10 7 9 11 16 7 2 1 7 2
31 29 16 10 3 2 18 2
4
3 3
4 6 1 18
2
4 1
1 2
1
1 1 1
% 96 100 50 55 30 41 95 10 13 21 4 2 6.5
5
10
15
20
monthi
Fig. 1. Time to progression for 20 patients. "•"Overall: 11 month* *-NC. 10.B monthi -*-CR • PR: 13.3 monthi
months
Fig. 2. Survival for 46 patients.
vomiting, and diarrhea were common but not limiting toxicities. Almost all patients experienced hypotension, which required a dose reduction of EL-2 in 6 patients. Substantial weight gain (> 5%) was not seen. Five patients developed some degree of anemia, but blood transfusions were not required. Thrombocytopenia occurred in 6 patients. Platelet transfusions were not given, but the dose of DTIC was reduced in one patient. Renal toxicities occurred in 10 patients, and required a dose reduction of IL-2 in 4 patients. Transient elevation of bilirubin and transaminases were experienced by a minority of patients. Treatment was stopped in one patient because of atrial fibrillation with hypotension. Three patients with central venous catheters developed sepsis, all with Staphylococcus aureus. The catheters were removed, and all of the patients recovered with appropriate antibiotic therapy. As is usual, no toxic effects could be related to the use of TA1. Immunological changes Table 3 summarizes the most important changes during treatment in hematological and immunological parameters. As expected, lymphopenia was recorded in all patients during IL-2 infusion. A significant increase in absolute number of lymphocytes and T cell subsets over baseline began to occur by day 15. T lymphocyte counts did not return to
744 Table 3. Changes during treatment in hematological and immunological parameters. Parameter
Lymphocytes CD3 CD4 CD8 DR+ CD16 CD25 CD19 TNFa B2M SIL-2R sCD4 sCD8 IgG IgM
Day 1
12
15
22
36
43
1902±635 1372±480 707±296 742±325 217±202 274±205 3311282 1711100 14.716.5 1.510.7 6231375 41170 4141139 10731392 120163
14021330' 1060151 2991 80b 8551140 5901157" 2421148 2141197 35141 b 166187' 3.811.6' 537113993' 26115 10131555' 8331202 114193
531412181' 391511844° 19451671' 220011648' 175111746" 10241653' 15511686" 2331240 54.5121.8' 3.411.2' 698313665' 921154 13481621' 9121267 2211133'
311111740' 225911207' 136311027 10811551" 5121519" 5431419" 8171866" 139190 29.0115.9' 2.411.2" 292513492' 1001206 5731175 b 14241529 2231111'
528914131' 384813265 264112909' 166711072' 169811557" 11881944" 235012737' 1631122 63.2129.4' 5.013.0' 771613716' 2031281" 10071406' 11101378 200180"
354611394" 257611098" 15711812' 11791552' 4831379' 7171484' 10811713' 1721275 25.7110.9' 2.210.7" 410616138' 50182 526 87" 13721412" 2721125'
Lymphocytes: cell/nL; TNFa: pg/ml; B2M: mg/L; s!L-2R, sCD4, sCD8: U/ml; IgG, IgM: mg/dl. In all parameters, statistical significance was calculated comparing day 1 vs. each subsequent day reported in the table. Values represent mean 1 SD. ' P-0.001. " P-0.05.
pre-treatment levels, and again rebounded with subsequent IL-2 administration. Cytokine levels increased during IL-2 administration. Except for TNFa, all of them returned to baseline levels by day 15. Serum levels of IL-1(3 and IL-4 showed considerable inter-patient as well as intra-cycle and inter-cycle variability. Among the soluble membrane molecules, the most relevant variation was observed in serum levels of sIL2R, with a 10-fold increase over baseline. To detect a cumulative effect of treatment, basal values of aL parameters were compared to the correspondent values on day 1 of the third cycle. A highly significant increase over baseline was observed in the following parameters: total lymphocytes, CD3, CD4, CD8, DR+, CD16, CD25, B2M, SIL-2R, TNFa and IgM. Comparison of changes in hematological and immunological parameters during treatment in responding and non-responding patients, showed significantly lower sCD8 levels in the latter group. An effort was also made to identify factors predictive of some clinical variables. The number of metastatic sites in this patient population ranged from 1 to 3. Pre-treatment serum levels of sCD4 were significantly lower in patients with 1 metastatic site in comparison to those with 2-3 sites of disease (P =• 0.05). Furthermore, basal values of CD4 and CD25 cells were significantly higher in non-responding than in responding patients (P - 0.04 and 0.005, respectively).
Discussion Experimental and clinical studies suggest that combined chemotherapy and IL-2 may have an additive or
synergistic effect against melanoma [7, 8]. Among the various mechanisms involved in this process [9], an appropriate potentiation of the cytokine modulatory effect may be of great importance. In order to achieve this goal, in the present study we have chosen TA1 because of its reported numerous immunomodulatory actions along with an absence of toxicity [4]. With a 36% overall response rate, the results reported here compare favorably with those of DTIC alone [1] and other recently published series of biochemotherapy involving DTIC and IL-2. In six phase I and II studies reported to date [10-15], responses have been observed in 34/145 patients for an overall response rate of 23% (range, 0% to 33%). Nevertheless, without a direct comparison with DTIC it is not possible to affirm that the combination of DTIC and TA1 + IL-2 is definitively better than DTIC alone, and the superiority of biochemotherapy over chemotherapy has not been formally established [16]. In this study, responses were seen in almost all tumor sites, although they occurred more frequently in patients with metastases limited to skin and soft tissue or lung. Progression of disease was generally apparent at the initially-involved sites. However, in 4 of the responding and 1 of the stabilized patients, CNS metastases developed while the disease was in remission or stable in extra-cerebral sites. This could explain the shorter time to progression and survival in CR and PR patients in comparison to those with stable disease. Treatment results did not seem to correlate with the extent of the disease, and we were encouraged by some responses in patients with visceral lesions or bulky disease. Although the median time to progression does not appear to be significantly different from that reported
745
by other investigators using chemotherapy or biochemotherapy regimens, the median survival of 11 months may be of some significance, inasmuch as it is similar or better than that observed in studies reporting a higher response rate [17, 18]. In this regard, the long duration of the stable disease observed in this study cannot be overlooked, and it is possible that biological agents may provide a survival advantage in the absence of disease progression. To better define the meaning of stable disease, all immunological data in responding and stable patients as a group were compared to those in progressive patients. Pre-treatment serum concentrations of sCD4 were significantly lower in the former group. During treatment, while sCD4 levels remained lower, sCD8 significantly increased in the same group of patients. No correlation was observed between sCD4 and sCD8 levels and the absolute number of circulating CD4 and CD8 cells in individual cases. Thus, it seems that patients who are most likely to benefit from treatment are those with 1 site of disease and lower baseline sCD4. These patients will manifest increased levels of sCD8 during treatment. In other words, activation of the CD8 T cell subset appears to be crucial. The side effects of this biochemotherapy combination were predominantly caused by IL-2. Despite the list of toxicities, treatment was tolerated reasonably well and an ICU support was not required. Nevertheless, central venous catheter-related problems may be troublesome to many patients, and subcutaneous administration of IL-2 should be considered in future studies [19]. It is noteworthy that the combined treatment resulted in neither overlapping toxicity nor interference with the hematological and immunological effects related to the use of TA1 + IL-2. T lymphocytes were generated at each treatment cycle. Furthermore, there was a continued stepwise expansion of the T cell subsets with repetitive cycles of TA1 + IL-2. These changes, as well as those in other parameters we have studied, are qualitatively similar to previously published data on patients treated with IL-2 ± chemotherapy [20,21]. However, a broad analysis of immunological parameters in patients receiving treatment protocols including thymic hormones has not been reported. Although the design of the present study did not allow us to assess the individual contribution of the agents used, the role of TA1 needs some comment. Immunological changes similar to those reported herein have been observed in a study with 5-fluorouracil and folinic acid combined with thymopentin and IL-2 in advanced colorectal cancer. In this tudy, significant changes in CD25, sIL-2R, sCD4, and IFN-gamma were observed in patients receiving thymopentin in comparison to a series of patients treated with the same regimen without thymic hormone (Lopez, unpublished data). In other studies [22] using chemotherapy with DL-2 alone, although there was a lymphocytosis after
each IL-2 administration, T lymphocyte counts returned to pre-treatment levels before the subsequent IL-2 treatment. Therefore, it is possible that thymic hormones may potentiate the action of IL-2. Although a comparison of the hematological and immunological changes between responders and non-responders showed no statistically significant difference, some differences in sCD4 and sCD8 levels could be observed when stable patients were evaluated together with those achieving a response. However, because of the small number of patients evaluated, the role of the released CD4 and CD 8 molecules will require further investigation. Although this treatment appears feasible and effective, more appropriate regimens are needed, and improvement in response rates to biochemotherapy in advanced melanoma may be expected. In experimental systems the immunomodulatory effect of thymic hormones appears to be related to the agent used as well as to the dosage and schedule of administration [23]. Our current studies are exploring some of the many ways in which thymic hormones can interact with cytokines as well as the optimal administration sequence of cytotoxic drugs and biological agents.
References 1. Houghton AN, Legha S, Bajorin DF. Chemotherapy for metastatic melanoma. In Balch CM, Houghton AN, Sober AJ et al. (eds): Cutaneous Melanoma. JB Lippincott: Philadelphia 1992; Chapter 39: 498-508. 2. Parkinson DR, Houghton AN, Hersey P et al. Biologic therapy for melanoma. In Balch CM, Houghton AN, Sober AJ et al. (eds): Cutaneous Melanoma. JB Lippincott: Philadelphia, 1992; Chapter 41: 522-41. 3. West WH, Tauer KW, Yannelli JR et al. Constant-infusion recombinant interleukin-2 in adoptive immunotherapy of advanced cancer. N Engl J Med 1987; 316: 898-905. 4. Kashmanian Oates K, Goldstein AL. Thymosin. In De Vita VT, Hellman S, Rosenberg SA (eds): Biologic Therapy of Cancer. JB Lippincott Philadelphia, 1991; Chapter 27: 705-18. 5. Mastino A, Favalli C, Grelli S et al. Combination therapy of thymosin a l potentiates the anti-tumor activity of interleukin-2 with cyclophosphamide in the treatment of the Lewis lung carcinoma in mice, bit J Cancer 1992; 50: 493-9. 6. Miller AB, Hoogstraten B, Staquet H et al. Reporting results of cancer treatment. Cancer 1981; 47: 207-14. 7. LoRusso PM, Polin L, Auckerman SL et al. Antitumor efficacy of interleukin-2 alone and in combination with adriamycin and dacarbazine in murine solid tumor systems. Cancer Res 1990; 50:4876-82. 8. Demchak PA, Mier JW, Robert NJ et al. A phase II pilot trial of interleukin-2 and high-dose cisplatin in patient with metastatic melanoma. J Clin Oncol 1991; 10:147-51. 9. Mitchell MS. Combining chemotherapy with biological response modifiers in treatment of cancer. J Natl Cancer Inst 1988; 80:1445-50. 10. Shiloni E, Pouillart P, Janssens J et al. Sequential dacarbazine chemotherapy followed by recombinant interleukin-2 in metastatic melanoma. A pilot multicentre phase I-II study. Eur J Cancer Clin Oncol 1989; 25 (Suppl 3}. 45-9. 11. Flaherty LE, Redman BG, Chabot GG et al. A phase I-II study of dacarbazine in combination with outpatient interleukin-2 in metastatic malignant melanoma. Cancer 1990; 65: 2471-7.
746 12. Papadopoulos NE, Howard JG, Murray JL et al. Phase II DTIC and Lnterleukin-2 trial for metastatic malignant melanoma Proc Am Soc Clin Oncol 1990; 9:277 (abstr). 13. Dillman RO, Oldham RK, Barth NM et al. Recombinant interleukin-2 and adoptive immunotherapy alternated with dacarbazine therapy in melanoma: A National Biotherapy Study Group trial. J Natl Cancer Inst 1990; 82:1345-9. 14. Stoter G, Aamdal S, Rodenhuis S et al. Sequential administration of recombinant human interleukin-2 and dacarbazine in metastatic melanoma: A multicenter phase II study. J Clin Oncol 1991; 9:1687-91. 15. Flaherty LE, Liu PY, Fletcher WS et al. Dacarbazine and outpatient interleukin-2 in treatment of metastatic malignant melanoma: Phase II Southwest Oncology Group trial. J Natl Cancer Inst 1992; 84: 893-4. 16. Thomson D, Adena M, McLeod GRC et al. Interferon a-2a (IFN) does not improve response or survival when added to dacarbazine (DTIC) in metastatic melanoma: Results of a multi-institutional australian randomised trial QMP8704. Proc Am Soc Clin Oncol 1992; 11: 343 (abstr.). 17. Richards JM, Mehta N, Ramming K et a). Sequential chemoimmunotherapy in the treatment of metastatic melanoma. J Clin Oncol 1992; 10:1338-43. 18. Lattanzi SC, Tosteson T, Maurer LH et al. Dacarbazine, cisplatin, and carmustine ± tamoxifen, in the treatment of patient with metastatic melanoma: Results of 5-year follow-up. Proc Am Soc Clin Oncol 1993; 12: 390 (abstr.). 19. Atzpodien J, Korfer A, Franks CR et al. Home therapy with recombinant interleukin-2 and interferon-a2b in advanced
human malignancies. Lancet 1990; 335:1509-12. 20. Lopez Hanninen E, Korfer A, Hadam M et al. Biological monitoring of low-dose interleukin-2 in humans: Soluble interleukin-2 receptors, cytokines and cell surface phenotypes. Cancer Res 1991; 50:6312-6. 21. Isacson R, Kedar E, Barak V et al. Chemo-immunotherapy in patients with metastatic melanoma using sequential treatment with dacarbazine and recombinant human interleukin-2: Evaluation of hematologic and immunologic parameters and correlation with clinical response. Immunol Lett 1992; 33: 127-34. 22. Redman BG, Flaherty L, Chou TH et al. Sequential dacarbazine/cisplatin and interleukin-2 in metastatic melanoma: Immunological effects of therapy. J Immunol 1991; 10: 147-51. 23. Papanastasiou M, Baxevanis CN, Papamichail M. Promotion of murine antitumor activity by prothymosin a treatment: I. Induction of tumoricidal peritoneal cells producing high levels of tumor necrosis factor a. Cancer Immunol Immunother 1992; 35:145-50. Received 20 December 1993; accepted 25 May 1994. Correspondence to: Dr. Massimo Lopez Divisione di Oncologia Medica II Istituto Regina Elena per lo Studio e la Cura dei Tumori Viale Regina Elena, 291 00161 Roma Italy
Original article Biochemotherapy with thymosin a l , interleukin-2 and dacarbazine in patients with metastatic melanoma: Clinical and immunological effects M. Lopez, S. Carpano, R. Cavaliere, L. Di Lauro, F. Ameglio,1 G. Vitelli, A. M. Frasca, P. Vici, F. Pignatti, M. Rosselli, G. Rasi2 & E. Garaci2 Regina Elena Institute for Cancer Research, Rome; '5. Gallicano Institute, Rome; 2'Tor Vergala' University, Rome, Italy
months and median survival was 11 months. Side effects were predominantly caused by IL-2. Treatment was tolerated Background: DTIC and interleukin-2 (IL-2), as single agents, reasonably well, and there was no overlapping toxicity or have a limited anti-tumor activity in patients with metastatic interference between chemotherapy and biotherapy. Baseline melanoma. Experimentally, thymosin a l (TA1) may modu- sCD4 levels seem to correlate to tumor burden. Patients late the action of IL-2. We investigated the clinical and im- benefiting from treatment had lower sCD4 and higher sCD8 munological effects of a combination with these three agents. than did progressing patients. Patients and methods: Forty-six patients with measurable Conclusions: The combination of DTIC + TA1 + IL-2 is metastatic melanoma were treated with DTIC 850 mg IV on active in the treatment of advanced melanoma, with acceptday 1, TA1 2 mg s.c. on days 4 to 7, and IL-2 18 MU/m2/d able toxicity. However, even more active regimens are needby continuous intravenous infusion on days 8 to 12. Cycles ed, and the interactions between thymic hormones and cytowere repeated every 3 weeks. kines should be further explored. Results: Objective responses were obtained in 15 (36%) of 42 evaluable patients (CI at 95%: 22%-50%). Two patients experienced complete responses, and stable disease was ob- Key words: advanced melanoma, biochemotherapy, DTIC, served in five. The median time to progression was 5.5 interleukin-2, thymosin a l Summary
Introduction
sin a l (TA1) may induce a wide variety of biological effects, including stimulation of NK activity, proliferative response of CD4+ cells to various antigens, production of several cytokines (IL-1, IL-2, IFN, TNFa), and expression of IL-2 receptors [4]. All these findings suggest that this agent may play a role in the modulation of the immune response against cancer. Interestingly, a recent experimental study by Mastino et al. reported a significantly improved survival of mice bearing Lewis lung carcinoma treated with combined cyclophosphamide (CTX), TA1 and IL-2 over that of mice receiving CTX, TA1 or IL-2 alone, or CTX plus a single immunomodulator [5]. Based on this clinical and experimental evidence, a phase II study was undertaken in patients with metastatic melanoma to evaluate the effectiveness and toxicity of dacarbazine combined with TA1 and CTV IL-2.
Metastatic melanoma is usually resistant to many forms of therapy. Only a small number of chemotherapeutic agents have demonstrated antitumor activity against this disease, and dacarbazine (DTIC) remains the most active single agent, with response rates of around 20%, usually of short duration and without impact on patient survival [1]. Combination chemotherapy has not proven to be clearly superior to DTIC alone [1]. In recent years, the availability of a number of cytokines along with a better understanding of the immune system has provided the opportunity to explore new approaches of treatment for melanoma. The most extensive experience has been gained with interferons (IFNj) and interleukin-2 (IL-2). When given alone, IL-2 was associated with response rates of up to 24%, which is similar to the rates observed when it was given in combination with LAK cells [2]. The regimen used in these trials included high-dose IL-2 by bolus injection, Patients and methods and was potentially associated with severe toxicity requiring intensive care unit (ICU) support for many pa- Patients tients. In an attempt to reduce toxicity and improve eligibility, patients were required to have a histologically or cytotherapeutic results, continuous intravenous (CIV) infu- For logically confirmed diagnosis of metastatic melanoma; bidimensionsion schedules have been developed [3]. ally measurable disease; age <75 years; World Health Organization There is an increasing body of evidence that thymo- (WHO) performance status (PS) <2; adequate bone marrow
742 (neutrophils > 1,500/uL; platelets > 100,000/(iL), renal (creatinine level < 1.5 mg/dL), and hepatic (bilirubin level < 1.5 mg/dL) function; no history or current evidence of cardiovascular disease; no infections requiring antibiotic therapy; no corticosteroids; and no brain metastases. Patients previously treated with chemotherapy, radiation therapy or interferon were not excluded provided that a recurrence had been documented and at least 4 weeks had elapsed since the end of the treatment. Forty-six patients were entered into the study and all were evaluable for toxicity. Four patients were considered not evaluable for response (refusal, 2; lost to follow-up, 1; incorrect diagnosis, 1). Patient characteristics are listed in Table 1. Twenty-five patients (54%) had >2 metastatic sites and 26 (56%) had visceral disease including lung in 22 patients and liver in 6 patients. Five patients had received prior chemotherapy, and six were given IFN in an adjuvant setting. All patients provided informed consent before admission to the trial.
1 week. If hematologic recovery was not complete by day 28, the DTIC dose was reduced by 50% for leukocyte counts between 2000/(iL and 2999/uL or platelets between 50,000/uL and 74,999/u.L. In the event of lower values, treatment was discontinued to allow recovery. Infusion of IL-2 was temporarily interrupted in the event of persistent severe hypotension, severe agitation or confusion, increase in serum bilirubin levels > 5 mg/dL, increase in serum creatinine levels >4,5 mg/dL, dyspnea at rest not reversed by binasal oxygen, bacterial sepsis, or other severe side effects. IL-2 doses were reduced by 50% if any of the following events had occurred in the previous cycle: hypotension that did not reverse within 48 h of stopping the agent, increase in serum creatinine levels >6 mg/dL, increase in serum bilirubin levels > 5 mg/dL or grade HI neurotoxicity. Treatment was definitively interrupted in instances of myocardial ischemia, grade IV neurotoxicity, increase in serum creatinine or bilirubin levels that failed to return to baseline values, or severe hypotension at reduced doses.
Treatment DTIC 850 mg/m2 IV over 30 minutes was given on day 1 with appropriate antiemetic therapy. Thymosin a l was kindly provided by Sclavo (Siena, Italy) in lyophilized vials containing 2 mg/vial, and was administered subcutaneously at a dose of 2 mg on days 4 to 7. IL-2 (Proleukin; Eurocetus Corporation, Amsterdam, The Netherlands) was given at a dose of 18 x 106 IU/mVd by a 24 h CTV on days 8 to 12. This required a central venous catheter, and the use of prophylactic antibiotics. Cycles were repeated every 3 weeks up to a maximum of 6 cycles. Treatment was discontinued in instances of disease progression, unacceptable toxicity or patient refusal. After completion of the treatment plan patients were followed until disease progression.
Dose modification No DTIC dose reduction was planned in instances of leukopenia (3000-3999 cells/uL) or thrombocytopenia (75,000-99,999 cells/ u.L) on days 21. Below these ranges, treatment was delayed by Table 1. Patient characteristics. Characteristics
No. of patients
Entered/evaluable Site of primary tumor Cutaneous Ocular Unknown Sex Male/Female Age (years) Median (range) Stage
46/42
in/iv Performance status (WHO) Median (range) Sites of disease Skin, soft tissues Lymph nodes Liver Lung Bone Adrenal Abdominal mass Spleen No. of metastatic sites 1/2/3/4
44 1 1 27/19 49.5 (22-74) 4/42
Supportive measures Paracetamol (650 mg x 6) or indomethacin (25 mg x 4-6) was used prophylactically to avoid fever, and ondansetron (8 mg x 3) was given to control nausea and vomiting. Hypotension was treated by intravenous colloid or crystalloid and/or dopamine (5-10 ng/kg/ min) as necessary.
Immunological studies Treatment-related immunological effects were evaluated in 22 patients (10 responders and 12 non-responders). Blood samples were taken before DTIC (day 1), TA1 (day 4) and IL-2 (day 8) administration, at the end of IL-2 infusion (day 12), and on the third day after completion of the treatment cycle (day 15). Lymphocyte phenotype (CD3, CD4, CD8, DR+, CD16, CD25, CD 19) was tested on fresh samples by FACS analysis, using commercially available monoclonal antibodies (Becton Dickinson, Mountain View, CA, USA). Serum aliquots were stored at -40 *C to permit all assays to be run in parallel simultaneously. Serum immunoglobulins (IgG, IgA, IgM) were determined by rate nephelometry (Beckman, Brea, CA, USA). IL-lp", IL-2, IL-6, IFN-gamma, and TNFa were measured by enzyme-amplified sensitivity immunoassays (Medgenix Diagnostics, Fleurus, Belgium), IL-4 was determined by sandwich enzyme immunoassay (Genzyme, Cambridge, MA, USA), beta-2-microglobulin (B2M) was determined by microparticle enzyme immunoassay (Abbott, Chicago, IL, USA), and soluble membrane molecules (sIL2R, sCD4, sCD8) were detected by sandwich enzyme immunoassays (T Cell Sciences, Cambridge, MA, USA). Data not normally distributed were assessed using the Anova and the Kruskal-Wallis tests. Statistical analysis was performed using the students t-test In the case of paired data a paired t-test was employed.
1(0-2) 19 20 6 22 3 2 4 1 21/20/1/4
Non-evaluable patients: refusal, 2; lost to follow-up, 1; incorrect diagnosis, 1.
Evaluations Before treatment, all patients underwent complete physical examination, chest x-ray, abdominal ultrasound or computer tomography (CT), and CT scans of the brain and chest. Hematologic and biochemical parameters were determined. The extent of disease was clinically assessed every 2 weeks. Lesions that were assessable by x-rays, ultrasonography or CT were checked every 2 cycles. Response and toxicity were evaluated according to WHO criteria (6]. For toxicities not reported in this system, the National Cancer Institute common toxicity criteria were used.
743 Response duration and survival were determined by the KaplanMeier method, and compared using the log-rank test. The Fisher's exact test was used to analyze differences in response rates and toxicity among subgroups of patients.
100
AJI pttianti: 5.6 months NC: 11.B monthi CR • PR: 4.7 months
Results 20 "
Response Objective responses were seen in 15 of 42 (36%; 95% confidence interval, 22% to 50%) patients. There were two complete responses (CR), in patients with skin and lung metastases, and 13 partial responses (PR). Five patients (12%) had stable disease (SD). Responses were seen in the following sites: lymph nodes, 3 of 20 (15%); skin and soft tissue, 7 of 19 (37%); lung, 10 of 22 (45%); liver, 1 of 6 (17%); abdominal masses, 1 of 4 (25%); adrenals, 1 of 2 (50%). No responses were seen in 3 patients with bone disease. Tumor regression was apparent after the first treatment cycle in 2 patients, and after the second and the third cycle in 3 and 10 patients, respectively. No correlation could be identified between patient pretreatment characteristics (previous treatment, performance status, visceral/non-visceral disease, number of metastatic sites, presence/absence of vitiligo) and likelihood of response to the biochemotherapy. After a maximum follow-up period of 29 months, the median time to progression (Fig. 1) for stabilized and responding patients is 5.5 months, for responding patients, 4.7 months, and 11.5 months for stable patients (CR + PR vs. SD: P = 0.03). The median survival is 11 months (Fig. 2), with 48% of patients surviving > 1 year. There is no statistical difference between responding patients and those with stable disease (P — 0.30). At present 9 patients are ah've. Toxicity All 46 patients were evaluable for toxicity, as shown in Table 2. Fatigue, fever, chills, skin rash, nausea and Table 2. Toxicity in 46 patients. Toxic effect
Fatigue Fever-chills Erythema-rash N&V Stomatitis Diarrhea Hypotension Anemia Thrombocytopenia Nephrotoxicity Hepatic Arrythmia Sepsis
WHO grade 0
1
2
2 3 23 21 32 27 2 41 40 36 44
10 10 7 9 11 16 7 2 1 7 2
31 29 16 10 3 2 18 2
4
3 3
4 6 1 18
2
4 1
1 2
1
1 1 1
% 96 100 50 55 30 41 95 10 13 21 4 2 6.5
5
10
15
20
monthi
Fig. 1. Time to progression for 20 patients. "•"Overall: 11 month* *-NC. 10.B monthi -*-CR • PR: 13.3 monthi
months
Fig. 2. Survival for 46 patients.
vomiting, and diarrhea were common but not limiting toxicities. Almost all patients experienced hypotension, which required a dose reduction of EL-2 in 6 patients. Substantial weight gain (> 5%) was not seen. Five patients developed some degree of anemia, but blood transfusions were not required. Thrombocytopenia occurred in 6 patients. Platelet transfusions were not given, but the dose of DTIC was reduced in one patient. Renal toxicities occurred in 10 patients, and required a dose reduction of IL-2 in 4 patients. Transient elevation of bilirubin and transaminases were experienced by a minority of patients. Treatment was stopped in one patient because of atrial fibrillation with hypotension. Three patients with central venous catheters developed sepsis, all with Staphylococcus aureus. The catheters were removed, and all of the patients recovered with appropriate antibiotic therapy. As is usual, no toxic effects could be related to the use of TA1. Immunological changes Table 3 summarizes the most important changes during treatment in hematological and immunological parameters. As expected, lymphopenia was recorded in all patients during IL-2 infusion. A significant increase in absolute number of lymphocytes and T cell subsets over baseline began to occur by day 15. T lymphocyte counts did not return to
744 Table 3. Changes during treatment in hematological and immunological parameters. Parameter
Lymphocytes CD3 CD4 CD8 DR+ CD16 CD25 CD19 TNFa B2M SIL-2R sCD4 sCD8 IgG IgM
Day 1
12
15
22
36
43
1902±635 1372±480 707±296 742±325 217±202 274±205 3311282 1711100 14.716.5 1.510.7 6231375 41170 4141139 10731392 120163
14021330' 1060151 2991 80b 8551140 5901157" 2421148 2141197 35141 b 166187' 3.811.6' 537113993' 26115 10131555' 8331202 114193
531412181' 391511844° 19451671' 220011648' 175111746" 10241653' 15511686" 2331240 54.5121.8' 3.411.2' 698313665' 921154 13481621' 9121267 2211133'
311111740' 225911207' 136311027 10811551" 5121519" 5431419" 8171866" 139190 29.0115.9' 2.411.2" 292513492' 1001206 5731175 b 14241529 2231111'
528914131' 384813265 264112909' 166711072' 169811557" 11881944" 235012737' 1631122 63.2129.4' 5.013.0' 771613716' 2031281" 10071406' 11101378 200180"
354611394" 257611098" 15711812' 11791552' 4831379' 7171484' 10811713' 1721275 25.7110.9' 2.210.7" 410616138' 50182 526 87" 13721412" 2721125'
Lymphocytes: cell/nL; TNFa: pg/ml; B2M: mg/L; s!L-2R, sCD4, sCD8: U/ml; IgG, IgM: mg/dl. In all parameters, statistical significance was calculated comparing day 1 vs. each subsequent day reported in the table. Values represent mean 1 SD. ' P-0.001. " P-0.05.
pre-treatment levels, and again rebounded with subsequent IL-2 administration. Cytokine levels increased during IL-2 administration. Except for TNFa, all of them returned to baseline levels by day 15. Serum levels of IL-1(3 and IL-4 showed considerable inter-patient as well as intra-cycle and inter-cycle variability. Among the soluble membrane molecules, the most relevant variation was observed in serum levels of sIL2R, with a 10-fold increase over baseline. To detect a cumulative effect of treatment, basal values of aL parameters were compared to the correspondent values on day 1 of the third cycle. A highly significant increase over baseline was observed in the following parameters: total lymphocytes, CD3, CD4, CD8, DR+, CD16, CD25, B2M, SIL-2R, TNFa and IgM. Comparison of changes in hematological and immunological parameters during treatment in responding and non-responding patients, showed significantly lower sCD8 levels in the latter group. An effort was also made to identify factors predictive of some clinical variables. The number of metastatic sites in this patient population ranged from 1 to 3. Pre-treatment serum levels of sCD4 were significantly lower in patients with 1 metastatic site in comparison to those with 2-3 sites of disease (P =• 0.05). Furthermore, basal values of CD4 and CD25 cells were significantly higher in non-responding than in responding patients (P - 0.04 and 0.005, respectively).
Discussion Experimental and clinical studies suggest that combined chemotherapy and IL-2 may have an additive or
synergistic effect against melanoma [7, 8]. Among the various mechanisms involved in this process [9], an appropriate potentiation of the cytokine modulatory effect may be of great importance. In order to achieve this goal, in the present study we have chosen TA1 because of its reported numerous immunomodulatory actions along with an absence of toxicity [4]. With a 36% overall response rate, the results reported here compare favorably with those of DTIC alone [1] and other recently published series of biochemotherapy involving DTIC and IL-2. In six phase I and II studies reported to date [10-15], responses have been observed in 34/145 patients for an overall response rate of 23% (range, 0% to 33%). Nevertheless, without a direct comparison with DTIC it is not possible to affirm that the combination of DTIC and TA1 + IL-2 is definitively better than DTIC alone, and the superiority of biochemotherapy over chemotherapy has not been formally established [16]. In this study, responses were seen in almost all tumor sites, although they occurred more frequently in patients with metastases limited to skin and soft tissue or lung. Progression of disease was generally apparent at the initially-involved sites. However, in 4 of the responding and 1 of the stabilized patients, CNS metastases developed while the disease was in remission or stable in extra-cerebral sites. This could explain the shorter time to progression and survival in CR and PR patients in comparison to those with stable disease. Treatment results did not seem to correlate with the extent of the disease, and we were encouraged by some responses in patients with visceral lesions or bulky disease. Although the median time to progression does not appear to be significantly different from that reported
745
by other investigators using chemotherapy or biochemotherapy regimens, the median survival of 11 months may be of some significance, inasmuch as it is similar or better than that observed in studies reporting a higher response rate [17, 18]. In this regard, the long duration of the stable disease observed in this study cannot be overlooked, and it is possible that biological agents may provide a survival advantage in the absence of disease progression. To better define the meaning of stable disease, all immunological data in responding and stable patients as a group were compared to those in progressive patients. Pre-treatment serum concentrations of sCD4 were significantly lower in the former group. During treatment, while sCD4 levels remained lower, sCD8 significantly increased in the same group of patients. No correlation was observed between sCD4 and sCD8 levels and the absolute number of circulating CD4 and CD8 cells in individual cases. Thus, it seems that patients who are most likely to benefit from treatment are those with 1 site of disease and lower baseline sCD4. These patients will manifest increased levels of sCD8 during treatment. In other words, activation of the CD8 T cell subset appears to be crucial. The side effects of this biochemotherapy combination were predominantly caused by IL-2. Despite the list of toxicities, treatment was tolerated reasonably well and an ICU support was not required. Nevertheless, central venous catheter-related problems may be troublesome to many patients, and subcutaneous administration of IL-2 should be considered in future studies [19]. It is noteworthy that the combined treatment resulted in neither overlapping toxicity nor interference with the hematological and immunological effects related to the use of TA1 + IL-2. T lymphocytes were generated at each treatment cycle. Furthermore, there was a continued stepwise expansion of the T cell subsets with repetitive cycles of TA1 + IL-2. These changes, as well as those in other parameters we have studied, are qualitatively similar to previously published data on patients treated with IL-2 ± chemotherapy [20,21]. However, a broad analysis of immunological parameters in patients receiving treatment protocols including thymic hormones has not been reported. Although the design of the present study did not allow us to assess the individual contribution of the agents used, the role of TA1 needs some comment. Immunological changes similar to those reported herein have been observed in a study with 5-fluorouracil and folinic acid combined with thymopentin and IL-2 in advanced colorectal cancer. In this tudy, significant changes in CD25, sIL-2R, sCD4, and IFN-gamma were observed in patients receiving thymopentin in comparison to a series of patients treated with the same regimen without thymic hormone (Lopez, unpublished data). In other studies [22] using chemotherapy with DL-2 alone, although there was a lymphocytosis after
each IL-2 administration, T lymphocyte counts returned to pre-treatment levels before the subsequent IL-2 treatment. Therefore, it is possible that thymic hormones may potentiate the action of IL-2. Although a comparison of the hematological and immunological changes between responders and non-responders showed no statistically significant difference, some differences in sCD4 and sCD8 levels could be observed when stable patients were evaluated together with those achieving a response. However, because of the small number of patients evaluated, the role of the released CD4 and CD 8 molecules will require further investigation. Although this treatment appears feasible and effective, more appropriate regimens are needed, and improvement in response rates to biochemotherapy in advanced melanoma may be expected. In experimental systems the immunomodulatory effect of thymic hormones appears to be related to the agent used as well as to the dosage and schedule of administration [23]. Our current studies are exploring some of the many ways in which thymic hormones can interact with cytokines as well as the optimal administration sequence of cytotoxic drugs and biological agents.
References 1. Houghton AN, Legha S, Bajorin DF. Chemotherapy for metastatic melanoma. In Balch CM, Houghton AN, Sober AJ et al. (eds): Cutaneous Melanoma. JB Lippincott: Philadelphia 1992; Chapter 39: 498-508. 2. Parkinson DR, Houghton AN, Hersey P et al. Biologic therapy for melanoma. In Balch CM, Houghton AN, Sober AJ et al. (eds): Cutaneous Melanoma. JB Lippincott: Philadelphia, 1992; Chapter 41: 522-41. 3. West WH, Tauer KW, Yannelli JR et al. Constant-infusion recombinant interleukin-2 in adoptive immunotherapy of advanced cancer. N Engl J Med 1987; 316: 898-905. 4. Kashmanian Oates K, Goldstein AL. Thymosin. In De Vita VT, Hellman S, Rosenberg SA (eds): Biologic Therapy of Cancer. JB Lippincott Philadelphia, 1991; Chapter 27: 705-18. 5. Mastino A, Favalli C, Grelli S et al. Combination therapy of thymosin a l potentiates the anti-tumor activity of interleukin-2 with cyclophosphamide in the treatment of the Lewis lung carcinoma in mice, bit J Cancer 1992; 50: 493-9. 6. Miller AB, Hoogstraten B, Staquet H et al. Reporting results of cancer treatment. Cancer 1981; 47: 207-14. 7. LoRusso PM, Polin L, Auckerman SL et al. Antitumor efficacy of interleukin-2 alone and in combination with adriamycin and dacarbazine in murine solid tumor systems. Cancer Res 1990; 50:4876-82. 8. Demchak PA, Mier JW, Robert NJ et al. A phase II pilot trial of interleukin-2 and high-dose cisplatin in patient with metastatic melanoma. J Clin Oncol 1991; 10:147-51. 9. Mitchell MS. Combining chemotherapy with biological response modifiers in treatment of cancer. J Natl Cancer Inst 1988; 80:1445-50. 10. Shiloni E, Pouillart P, Janssens J et al. Sequential dacarbazine chemotherapy followed by recombinant interleukin-2 in metastatic melanoma. A pilot multicentre phase I-II study. Eur J Cancer Clin Oncol 1989; 25 (Suppl 3}. 45-9. 11. Flaherty LE, Redman BG, Chabot GG et al. A phase I-II study of dacarbazine in combination with outpatient interleukin-2 in metastatic malignant melanoma. Cancer 1990; 65: 2471-7.
746 12. Papadopoulos NE, Howard JG, Murray JL et al. Phase II DTIC and Lnterleukin-2 trial for metastatic malignant melanoma Proc Am Soc Clin Oncol 1990; 9:277 (abstr). 13. Dillman RO, Oldham RK, Barth NM et al. Recombinant interleukin-2 and adoptive immunotherapy alternated with dacarbazine therapy in melanoma: A National Biotherapy Study Group trial. J Natl Cancer Inst 1990; 82:1345-9. 14. Stoter G, Aamdal S, Rodenhuis S et al. Sequential administration of recombinant human interleukin-2 and dacarbazine in metastatic melanoma: A multicenter phase II study. J Clin Oncol 1991; 9:1687-91. 15. Flaherty LE, Liu PY, Fletcher WS et al. Dacarbazine and outpatient interleukin-2 in treatment of metastatic malignant melanoma: Phase II Southwest Oncology Group trial. J Natl Cancer Inst 1992; 84: 893-4. 16. Thomson D, Adena M, McLeod GRC et al. Interferon a-2a (IFN) does not improve response or survival when added to dacarbazine (DTIC) in metastatic melanoma: Results of a multi-institutional australian randomised trial QMP8704. Proc Am Soc Clin Oncol 1992; 11: 343 (abstr.). 17. Richards JM, Mehta N, Ramming K et a). Sequential chemoimmunotherapy in the treatment of metastatic melanoma. J Clin Oncol 1992; 10:1338-43. 18. Lattanzi SC, Tosteson T, Maurer LH et al. Dacarbazine, cisplatin, and carmustine ± tamoxifen, in the treatment of patient with metastatic melanoma: Results of 5-year follow-up. Proc Am Soc Clin Oncol 1993; 12: 390 (abstr.). 19. Atzpodien J, Korfer A, Franks CR et al. Home therapy with recombinant interleukin-2 and interferon-a2b in advanced
human malignancies. Lancet 1990; 335:1509-12. 20. Lopez Hanninen E, Korfer A, Hadam M et al. Biological monitoring of low-dose interleukin-2 in humans: Soluble interleukin-2 receptors, cytokines and cell surface phenotypes. Cancer Res 1991; 50:6312-6. 21. Isacson R, Kedar E, Barak V et al. Chemo-immunotherapy in patients with metastatic melanoma using sequential treatment with dacarbazine and recombinant human interleukin-2: Evaluation of hematologic and immunologic parameters and correlation with clinical response. Immunol Lett 1992; 33: 127-34. 22. Redman BG, Flaherty L, Chou TH et al. Sequential dacarbazine/cisplatin and interleukin-2 in metastatic melanoma: Immunological effects of therapy. J Immunol 1991; 10: 147-51. 23. Papanastasiou M, Baxevanis CN, Papamichail M. Promotion of murine antitumor activity by prothymosin a treatment: I. Induction of tumoricidal peritoneal cells producing high levels of tumor necrosis factor a. Cancer Immunol Immunother 1992; 35:145-50. Received 20 December 1993; accepted 25 May 1994. Correspondence to: Dr. Massimo Lopez Divisione di Oncologia Medica II Istituto Regina Elena per lo Studio e la Cura dei Tumori Viale Regina Elena, 291 00161 Roma Italy