Chemo-immunotherapy in patients with metastatic melanoma using sequential treatment with decarbazine and recombinant human interleukin-2: evaluation of hematologic and immunologic parameters and correlation with clinical response

Immunology Letters, 33 (1992) 127-134

0165 - 2478 / 92 / $ 5.00 © 1992 ElsevierSciencePublishers B.V. All rights reserved IMLET 01807

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 Rut Isacson a, Eli Kedar b, Vivian Barak a, Zulma Gazit b, Oded Yurim c, Inna Kalichman a, Hannah Ben-Bassat d, Shoshana Biran a, Michael Schlesinger e, Christopher R. Franks f, G e r d a J. R o e s t f, Peter A. P a l m e r f a n d Eitan Shiloni c aDepartment of Oncology, bLautenberg Center for General and Tumor Immunology, eDepartment of Surgery B, dDepartment of Experimental Surgery, eDepartment of Experimental Medicine, Hadassah University Hospital, Jerusalem, Israel; and fEuroCetus B.V., Amsterdam, The Netherlands

(Received 13 January 1992;revision received28 April 1992; accepted 29 April 1992)

1. Summary We have treated 18 patients with metastatic malignant melanoma (MM) with high-dose IL-2 administered by continuous iv infusion in combination with dacarbazine (DTIC), and correlated the clinical response with various hematologic and immunologic parameters. Two regimens differing in the sequence of treatment were employed, and 1-6 treatment cycles were given, depending on patient response. Two patients had a complete response (CR, 4 6 + m , 14m), two patients a partial response (PR, 16m,6m), one a minimal response and four had a stable disease lasting 2-7 months, thus the response rate (CR + PR) was 22%. None of the following parameters, tested prior to initiation of the therapy Key words: Melanoma; Dacarbazine; Interleukin-2 Correspondence to: Dr. Rut Isacson, Department of Ontology, Hadassah UniversityHospital, Jerusalem 91020, Israel. Abbreviations: MM, malignant melanoma; NK, natural killer;

LAK, lymphokine-activated killer; IL-2, interleukin-2; iv, intravenous; TNF~t, tumor necrosis factor ct; WBC, white blood cells; PBMNL, peripheral blood mononuclear leukocytes.

and 1-2 days after termination of each course of IL-2, correlated with the clinical response: WBC counts (total and differential), levels of blood CD4 and CD8 T cells, N K cells, monocytes and B cells, production of IL-1 and IL-1 inhibitor by monocytes, responsiveness to 3 mitogens, N K / L A K cell activity, and serum levels of IL-l~t, IL2, soluble IL-2 receptor, and TNF~. The only prognostic parameter was the greater increase in the level of IL-2 receptor (Tac)-bearing lymphocytes in the responding patients after 1-3 cycles of IL-2. \The data suggests that non-specific immune parameters have no prognostic value for patients undergoing IL-2-based immunotherapy.

2.

Introduction

Metastatic malignant melanoma (MM) carries a dismal prognosis. Dacarbazine (DTIC) is the most effective chemotherapeutic agent in M M with a response rate of 15-25%; most of these responses are, however, partial and of a short duration [1]. Considering the failure of current chemotherapy to cure or significantly prolong survival of patients with disseminated MM, the search for new therapeutic strategies is mandatory. Immuno127

therapy with interleukin-2 (IL-2), alone or together with IL-2-activated autologous lymphocytes (lymphokine-activated killer (LAK) cells), gave a response rate of approx. 20%, with 5-10% of the patients achieving a complete response in some studies [2-4]. The synergistic therapeutic effects obtained in experimental tumor systems with several chemotherapeutic agents in combination with IL-2 [5-9], stimulated the application of similar regimens in clinical trials. In this phase II study, which is part of a European multicenter study, supported by EuroCetus, D T I C and recombinant human IL-2 (given by continuous iv infusion) were sequentially administered to 18 M M patients. Preliminary reports on clinical responses and toxicity in some of the patients included in our study were recently published [10-12]. We present here the analysis of the hematologic and immunologic parameters tested prior, during, and post-therapy. We report our results separately, because such a broad analysis has not been performed by the other centers included in this study.

repeated at 5 weeks. Maintenance therapy was scheduled 3 weeks after the completion of induction treatment, consisting of IL-2 18 x 106 IU/ m2/24 h for 5 days, alternating with DTIC 850 mg/m 2 i.v. every 3 weeks for 18 weeks. A maximum of 6 cycles of IL-2 were scheduled.

3.2.2. Protocol H (13patients) DTIC 850 mg/m 2 was given by short i.v. infusion on day 1; IL-2 18 × 106 IU/m2/24 h was administered by continuous i.v. infusion for 5 consecutive days (days 4-9). A maximum of 6 treatment cycles with DTIC and IL-2 were scheduled: the first two at 13-day intervals and the next four at 20-day intervals. Clinical response was evaluated every 2 cycles of IL-2. Treatment was discontinued upon disease progression. Further details regarding treatment are given in references [10-12]. The IL-2 ( > 9 7 % pure, 18 x 106 IU/mg) was kindly supplied by EuroCetus B.V., Amsterdam, The Netherlands. 3.3.

3.

Blood samples

Materials and Methods

Eighteen previously untreated patients (11 males, 7 females) with a median age of 40 years (range: 18-64) and a Karnofsky status >/80% were treated. All patients had histologically proven disseminated melanoma with measurable disease (Table 1). Patients were selected according to the criteria defined previously [10-12]. Treatment was given between February 1988 and April 1990. All 18 patients were evaluable for analysis.

Blood (15-25 ml) was drawn with and without heparin immediately prior (baseline), during IL-2 infusion, and 1 or 2 days after each course of IL2. Blood samples were also taken from some patients 4-6 days after IL-2 and 2-4 days after DTIC. Serum aliquots were stored at - 7 0 ° C . Peripheral blood mononuclear leukocytes (PBMNL) were separated by a Ficoll-Hypaque gradient. For cryopreservation, cell aliquots were suspended in 10% DMSO, 20% fetal calf serum and 70% RPMI 1640 medium and stored at - 180°C for 2 weeks up to 12 months.

3.2.

3.4.

3.1.

Patient population

Treatment

Two different regimens of DTIC and IL-2 were employed.

3.2.1. Protocol I (5patients) Two induction cycles of IL-2 18 × 10 6 IU (equivalent to 3 × 106 Cetus Units)/m2/24 h were given by continuous i.v. infusion on days 1-5 and 12-16. DTIC 850 mg/m 2 was given by short i.v. infusion (20-30 min) on day 26. This cycle was 128

Assessment of hematologic parameters and immune functions in vitro

White blood cell (WBC) counts were performed on fresh samples using Coulter counting. Differential counts were done both by Coulter and on Giemsa stained smears. Cell markers, immune functions and cytokine serum levels were assayed on fresh or frozen samples, or on both (see Results). Except for N K / L A K activity which was considerably reduced by cryopreservation

TABLE 1 Patient characteristics Patient No. a

Age

Sex

Disease sites

IL-2 cycles given

D T I C cycles % of IL-2 given scheduled dose given

Response duration (months) b

1 2 3 4 5

37 50 64 49 37

M M M M F

Lung, liver, L N e Liver, spleen Lung, liver, L N SC d nodules Liver, L N

2 4 2 6 2

2 1 5 1

100 100 100 90 100

M R (5) P R (6) P C R (46 + ) P

6 7 8 9 10 11 12 13 14 15

34 42 41 22 39 18 33 55 40 42

M F M F M F F F M

6 4 2 2 4 2 6 2 2

6 4 2 2 4 2 6 2 2

80 70 85 90 70 100 85 90 100

S (7) S (2) P P

16 17 18

38 24 50

LN, SC nodules Lung Lung, liver LN Spleen Lung, liver, SC nodules Mediastinal mass SC nodules LN Adrenal, lung, SC nodules Lung, liver LN Liver

2 6 1 6

2 6 1 4

95 85 100 60

M M M F

CR (14) P

s (5) P P P P R (16) P

s (7)

apatients Nos. 1-5-were treated according to protocol I, patients Nos. 6-18 according to protocol II (see Materials and Methods). bCR, completeresponse; PR, partial response; MR, minimal response; S, stable; P, progression. ¢LN, lymph nodes. dSC, subcutaneous. (see Results), no significant differences in the other parameters tested were noted in preliminary experiments between fresh and cryopreserved samples. Using cryopreserved samples permitted several assays to be run in parallel simultaneously on samples taken at different times during treatment, thereby avoiding day-to-day variations in experimental conditions. Each frozen sample was tested in 2-4 separate experiments and the results represent the means. PBMNL phenotype was tested by standard immunofluorescence techniques and FACS analysis, using the following monoclonal antibodies (Becton-Dickinson, Mountain view, CA): Leu4 (CD3) for total T cells; Leu2 (CD8), Leu3 (CD4), B4 (CD19) for B cells; Ia (HLA-DR), Leu M3 (CD14) for monocytes; Leul 1 (CD16) for FCy receptor-bearing cells; and Tac (CD25) for IL-2 receptor-bearing cells. Results are also expressed in stimulation index = peak level during treatment/ baseline level. Levels of cytokines (IL-I~, IL-2, TNFct) and

soluble IL-2 receptor in patient sera were assayed by enzyme-linked immunosorbent assays (ELISA) (T Cell Sciences, Cambridge, MA). IL-2 was also determined by a bioassay, using a murine T cell line (CTLL-2) in a 48 h [3H]thymidine uptake assay, with results similar to those obtained by ELISA. In vitro secretion of IL-0t and IL-1 inhibitor by freshly obtained blood monocytes during a 24 h incubation was tested as described previously [13]. Results are expressed in stimulation index (see above). Response to the mitogens phytohemagglutinin (PHA), concanavalin A (con A) and pokeweed mitogen (PWM) was tested as described previously [14]. Cytotoxic activity of PBMNL was assayed at 1:1-100:1 effector/target cell ratios in a 4 h 51Cr-release assay against the NK-sensitive target K562 cells and the NK-resistant, LAK-sensitive target SW 48 human colorectal carcinoma cell line, as previously described [15]. Cells of several patients were tested in parallel against the LAKsensitive A375 human melanoma cell line, with re129

suits similar to those obtained with the SW48 cell line. Results are expressed in mean lytic unit (LU)/106 cells, where 1 LU = number of effector cells required to lyse 20% target cells, using 3000 labeled target cells/well. 3.5.

Statistical analysis

Differences between the responding and nonresponding patient groups for each parameter were tested for significance (P <0.05) using a two-tailed Student's t-test. 4.

4.1.

Results

Patient characteristics and clinical response

Table 1 shows the patient characteristics, the treatment details and the clinical responses. Eleven patients received 90-100% of the scheduled IL-2 dose, 7 patients received 60-85% of the planned dose due to toxicity. Of the 18 patients, two achieved a complete response ( > 4 6 months; 14 months) and two a partial response (16 and 6 months), thus the response rate ( C R + P R ) was 22%. One patient had a minimal response and four patients had a stable disease for 2-7 months. 4.2.

Hematologic parameters

As reported by other investigators, a marked lymphopenia was observed during IL-2 infusion, as compared with baseline levels (mean lymphocyte number 610 vs. 1925//~1). Two days after completion of IL-2 administration there was a marked increase in WBC counts (rebound leukocytosis). Peak levels were noted usually after the first two IL-2 cycles (mean baseline 7916/pl, mean post-IL-2 25083//A, range 8000-54000). This was accompanied by lymphocytosis (mean 8926/#1 vs. baseline 1925//~1). A marked eosinophilia (range 9-42%) was seen in all patients during and 2 days after IL-2 administration (mean baseline 240/#1, during IL-2 1480/#1, post-IL-2 1900//A). There was no correlation between the clinical response and lymphocytosis or eosinophilia tested 2 days after IL-2 infusion nor at any other time points during treatment (data not shown). 130

4.3.

Blood cell subpopulations

The PBMNL subsets of 8 non-responders and 4 responders were analyzed prior to and 2 days after each course of IL-2 (Table 2). The following alterations in the percentage of the various subpopulations during treatment were noted: (a) no significant change in Leu 4 + (total T cells) with a slight decrease in Leu2+ (CD8) T cells in both groups; (b) a significant increase in H L A - D R + cells in both groups; (c) a decrease in B4 + (CD19) B cells in both groups; (d) a slight increase or a decrease in L e u l l + (CD16) cells in non-responders and responders, respectively; and (e) a marked increase (mean 10-24-fold, relative to pretreatment level) in Tac + (CD 25, IL-2 receptor) cells in both groups. Peak levels of the various subsets were seen following the first or second cycle of IL-2 in the majority of patients. The only parameter that significantly differed between the two patient groups was the higher increase in TAC + cells in the responder group following IL-2 administration (P = 0.02). 4.4.

NK and LAK cell activity

To avoid test variations, all samples of cryopreserved PBMNL from a single patient were tested simultaneously; cells collected at each time point were tested in at least t w o separate experiments. It should be noted that the cytotoxic activity of the frozen cells was 30-40% lower (calculated in LU/106 cells) as compared with that of the same cell sample tested before cryopreservation. As shown in Table 3, a marked increase (/> 2 x the basal activity) of N K activity was observed in 13/ 18 (72%) of the patients (4/5 in protocol I and 9/ 13 in protocol II). L A K activity was increased in 9/18 patients, 5/5 in protocol I and 4/13 in protocol II. No correlation was found between N K / L A K cell activation and clinical response. Three out of the 4 responding patients had a poor L A K cell activity both prior to and after IL-2 treatment; in contrast, one patient (No. 3) with a very strong cytotoxic activity showed a rapidly progressive disease. In the majority of the patients maximal potentiation of N K / L A K cell activity was observed 1-2 days after the first or second IL-2 cycle. Blood cells obtained at other times in

TABLE 2 Analysis of cell surface markers Marker a

Non-responders (n = 8)

Responders (n = 4)

Mean (+ SD)

Range

Mean (_ SD)

Range

Leu 4 (CD3)

B I

74.3 (6.9) 1.0 (0.2)

62.6-84.0 0.6-1.2

76.9 (4.9) 1.1 (0.0)

72.0-83.7 1.0-1.2

Leu 3 (CD4)

B I

45.6 (5.8) 1.2 (0.2)

38.9-58.3 0.7-1.3

48.0 (11.0) 1.3 (0.2)

34.~61.2 1.0-1.4

Leu 2 (CD8)

B I

31.0 (8.2) 0.8 (0.2)

22.6-48.5 0.6-1.2

32.1 (5.7) 0.8 (0.3)

25.0-39.0 0.4-1.1

Ia (HLA-DR)

B I

15.6 (2.9) 3.1 (1.7)

10.5-18.7 1.2-3.6

14.6 (4.1) 3.2 (0.1)

10.5-20.2 3.1-3.3

Leu M3 (CD14)

B I

4.4 (2.4) 1.0 (0.7)

1.3-8.9 0.1-2.0

B4 (CD19)

B I

8.6 (2.9) 0.7 (0.6)

5.6-13.7 0.2-1.9

5.1 (2.7) 0.4 (0.4)

3.5-7.7 0.2-1.0

Leu 11 (CD16)

B I

5.4 (2.8) 1.7 (1.1)

2.5-9.5 0.2-3.7

8.8 (2.9) 0.9 (0.4)

7.2-13.9 0.2-1.3

Tac (CD25)

B I

2.0 (1.1) 10.7 (5.3)b

0.5-3.5 4.6-19.0

0.9 (0.4) 24.0 (5.9)

0.3-1.3 18.0-32.0

Not tested

aB, Baseline level (percentage of total PBMNL); I, Index = peak level 2 days post IL-2/baseline level. Tested on both fresh and cryopreserved PBMNL with similar results, bSignificant difference between responders and non-responders (P = 0.02). the course o f treatment also failed to show a significant difference in cytotoxic activity between the responding and non-responding patients (data not shown). The mitogenic response o f fresh P B M N L to P H A , C o n A and P W M prior to and 1-2 days after IL-2 was quite variable, with a tendency to reduced responsiveness post-treatment. There was no correlation between the mitogenic response and the clinical response (data n o t shown).

4.4. Cytokines The serum levels o f IL-I~, IL-2, T N F ~ and soluble IL-2 receptor were determined before and 2 days after completion o f IL-2 infusion in 12 nonresponders and the 4 responders (Table 4). In addition, serum IL-2 levels were assayed in all 18 patients during IL-2 infusion, and blood m o n o -

cytes collected 2 days after IL-2 from 6 non-responders and 3 responders were tested for in vitro p r o d u c t i o n o f I L - I ~ and IL-1 inhibitor. Post-treatment serum levels o f soluble IL-2 receptor and T N F ~ were significantly higher (P < 0.05) than baseline levels in both groups. There was no significant increase in I L - I ~ (serum level and m o n o c y t e production) and IL-1 inhibitor following IL-2 administration (data not shown). The serum IL-2 levels during continuous infusion ranged f r o m 20-120 IU/ml. N o n e o f these parameters, tested b o t h pre- and post-treatment, showed statistically significant differences between the responding and non-responding patient groups. Interestingly, the pretreatment T N F ~ levels in 6 o f 12 non-responders were markedly higher than those in the 4 responders, and the levels increased after IL-2 treatment in all 4 responders but only in 6/12 non-respon131

TABLE 3 N K and LAK cell activity before and after IL-2 treatment a Patient

Protocol

Pre-treatment NK

Peak activity post-treatment b LAK

NK

LAK

1 2c 3 4c 5

1

18 27 58 25 26

<1 < 1 48 < 1 < 1

81 163 8300 26 63

(2) (2) (2) (5) (2)

6 7 8 9 10c

2

13 < 1 57 2 86

31 < 1 8 < 1 4

67 23 63 7 57

(2) (1) (1) (I) (2)

11

8

< 1

12 13 14 15 16c 17 18

55 < 1 43 5 11 145 11

<1 2 1 < 1 <1 3 4

44 (2)

157 17 55 45 14 360 77

(3) (2) (2) (1) (1) (1) (1)

91 6 2134 2 2

(2) (2) (2) (5) (2)

58 (1) 13 (3) 8 (1) < 1 (1) < 1 (2) < 1 (2)

<1 (3) 2 (2) 6 (2) < 1 (1) <1 (2) < 1 (1) 7 (2)

aTests were carried out with cryopreserved P B M N L obtained 1-2 days after completion o f each IL-2 cycle. Results presented in LU/106 cells and are means o f 2-3 repeated experiments, bin parentheses, n u m b e r o f IL-2 courses after which peak cytotoxic activity was observed. CPatients w h o achieved an objective clinical response.

ders. Except for a higher proportion of patients showing increased LAK cell activity among those included in protocol I, the two treatment protocols were comparable in terms of therapeutic efficacy and changes in hematologic/immunologic parameters. These conclusions are based, however, on a small number of patients. 5.

Discussion

The objectives of the present study were: (a) to evaluate in MM the therapeutic efficacy of a combination treatment with DTIC and high-dose IL2, and (b) to assess a broad spectrum of hematologic parameters and immune functions in an attempt to identify predictive factors of clinical response. The 22% response rate (2CR, 2PR/18 patients) in this study is similar to that reported in other studies with MM that included low- or high-dose IL-2 in combination with either DTIC [10132

12,16,17], low-dose cyclophosphamide [18], or doxorubicin [19], or with high-dose IL-2 + / LAK cell administration without chemotherapy [2-4]. This response rate is slightly higher than those obtained in other studies with high-dose IL-2 + / - LAK cells [20,21]. In view of the impressive therapeutic effects of chemotherapy combined with IL-2 in animal tumor models [5-9], the results are rather disappointing. However, the studies in animals show that the therapeutic efficacy of chemoimmunotherapy is highly dependent on the types of the chemotherapeutic agents used and the treatment schedule, in particular, the sequence and time interval between the administration of chemotherapy and IL-2 [22-24]. It is therefore possible that the two combination regimens employed by us were not optimal. In several trials attempts were made to correlate the clinical response to IL-2-based immunotherapy with changes in hematologic and immunologic parameters. In the majority of studies

TABLE 4 Serum levels of IL-la, IL-2, TNF0t and soluble IL-2 receptor pre- and post-IL-2 infusiona Parameter b

Non-Responders (n = 12)

Responders (n = 4)

Mean (___SD)

Mean ( + SD)

Range

Range

IL-I (pg/ml)

B I

329 (304) 1.1 (0.3)

190-1280 0.6-1.7

199 (61) 1.1 (0.3)

121-294 0.8-1.6

IL-2 (IU/ml)

B I

6.1 (2.9) 3.8 (4.4)

1.3-11.1 0.3-16.8

7.3 (4.4) 1.4 (1.2)

3.0-14.0 0.7-3.2

TNF~t (pg/ml)

B I

4.2 (4.4) 3.6 (3.5)

1.0-13.0 0.2-12.0

1.0 (0) 5.7 (2.9)

3.0-10.0

B I

1067 (357) 17.5 (6.1)

640-1750 6.2-26.2

835 (213) 24.2 (5.0)

590-1150 14.9-30.0

S.IL-2R (U/ml)

aTested on frozen serum samples taken before and 2 days after termination of IL-2 administration, bB, baseline level; I, index = peak level/baseline level. In the majority of cases peak levels were found after 1 or 2 IL-2 cycles.

such a correlation was not seen [25-27]. In only a few studies was the clinical response associated with the degree of lymphocytosis [28], the level of LAK cell activity [18], or the serum concentrations of IL-1 and TNF0t post-treatment [29]. In our study, as well as in another study [30], the only prognostic factor was the greater elevation in the proportion of Tac + cells after IL-2 administration in the responding patients. The biological significance of this increase in the responders is unclear. The changes in hematologic and immunologic parameters studied by us are similar to previously published data on patients treated with IL-2 + / LAK cells. However, a broad analysis of immune parameters in patients treated with DTIC (or other chemotherapy) and IL-2 has not been reported. Extensive studies in animal tumor models demonstrated that a specific T cell response is crucial for tumor rejection [31,32]. It is therefore not surprising that the above non-specific parameters usually do not correlate with the clinical response to immunotherapy. In the few cases where the in vitro tests correlated with the clinical response (see above), this correlation was found only posttreatment. Thus, the assessment of baseline levels of the variables tested by us cannot provide a meaningful indication for clinical response, and

therefore these parameters are useless for the selection of patients who are likely to respond to IL-2-based immunotherapy. Since only a small proportion (~<25%) of MM patients currently benefit from immunotherapy, it would be of great importance to identify, prior to initiation of treatment, specific immune factors which may be of prognostic value [31,33]. At present, the only specific immune parameter which seems to correlate with a favorable clinical course, post-surgery, in some patient groups (mainly with localized lung tumor or soft tissue sarcoma) is the ex vivo autologous tumor killing by freshly obtained blood lymphocytes [34-36]. Other assays to be considered are: the autologous lymphocyte stimulation (DNA synthesis) in mixed lymphocyteautologous tumor cell cultures, skin tests (DTH response) to autologous tumor cells or extracts, and analysis of melanoma specific antigens, MHC class I/II antigens and adhesion molecules on the tumor cells. HLA phenotyping may also be valuable, as MM patients who are HLA-A2, B12s and C3, particularly in combination, usually respond better to active specific immunotherapy with a tumor cell vaccine [37]. In conclusion, it can be expected that the response rate to chemoimmunotherapy in MM can be improved, provided there is the introduction of more stringent criteria for patient selection, and 133

the refinement of the regimens and treatment schedule.

Acknowledgements We gratefully acknowledge the support of the Israel Cancer Association, the Society of Research Associates of the Lautenberg Center, the Concern Foundation of Los Angeles, the Wakefern/ShopRite Endowment for Basic Research in Cancer Biology and Immunology, and the Harold B. Abramson Memorial Fund.

References [1] Koh, H.K. (1991) N. Engl. J. Med. 325, 171. [2] Rosenberg, S.A, Lotze, M.T., Yang, J.C., Aebersold, P.M., Linehan, W.M., Seipp, C.A. and White, D.E. (1989) Ann. Surg. 210, 474. [3] Gaynor, E.R., Weiss, G.R., Margolin, K.A., Aronson, F.R., Sznol, M., Demchak, P., Grima, K.M., Fisher, R.I., Boldt, D.H., Doroshow, J.H., Bar, M.H., Hawkins, M.J., Meir, J.W. and Caliendo, G. (1990) J. Natl. Cancer Inst. 82, 1397. 141 Dillman, R.O., Oldham, R.K., Tauer, K.W., Orr, D.W., Barth, N.M., Blumenschein, G., Arnold, J., Birch, R. and West, W.H. (1991) J. Clin. Oncol. 9, 1233. [5] Silagi, S. and Schaefer, A.E. (1986) J. Biol. Response Modif. 5, 411. [6] Kedar, E., Ben-Aziz, R. and Shiloni, E. (1988) Israel J. Med. Sci. 24, 494. [7] Wiltrout, R.H. and Salup, R.R. (1988) Prog. Exp. Tumor Res. 32, 128. [8] Papa, M.Z., Yang, J.C., Vetto, J.T., Shiloni, E., Eisenthal, A. and Rosenberg, S.A. (1988) Cancer Res. 48, 122. [9] LoRusso, P.M., Aukerman, S.L., Polin, L., Redman, B.G., Valdivieso, M., Biernat, L. and Corbett, T.H. (1990) Cancer Res. 50, 5876. [10] Shiloni, E., Pouillart, P., Janssens, J., Splinter, T., Di Peri, T., Symann, M., Roest, G.J., Palmer, P.A. and Franks, C.R. (1989) Eur. J. Cancer Clin. Oncol. 25, Suppl. 3, $45. [11] Stoter, G., Shiloni, E., Aamdal, S., Ceton, F.J., Iacobelli, S., Bijman, J.T., Palmer, P., Franks, C.R. and Rodenhuis, S. (1989) Eur. J. Cancer Clin. Oncol. 25, Suppl. 3, $41. [12] Stoter, G., Aamdal, S, Rodenhuis, S., Cleton, F.J., Iacobelli, S., Franks, C.R., Oskam, R. and Shiloni, E. (1991) J. Clin. Oncol. 9, 1687. [13] Barak, V., Treves, A.J., Yanai, P., Halperin, M., Wasserman, D., Biran, S. and Braun, S. (1986) Eur. J. lmmunol. 16, 1449. [14] Kedar, E., Rezai, A.R., Giorgi, J., Gale, R.P., Champlin, R.E., Mitsuyasu, R.T. and Fahey, J.L. (1988) Natl. Immun. Cell Growth Regul. 7, 13. [15] Kedar, E., Ikejiri, B., Bonnard, G.D. and Herberman,

134

R.B. (1982) Eur. J. Cancer Clin. Oncol. 18, 991. [16] Dillman, R.O., Oldham, R.K., Barth, N.M., Birch, R., Arnold, J. and West, W.H. (1990) J. Natl. Cancer Inst. 82, 1345. [17] Flaherty, L., Redman, B.G., Chabot, G., Martino, S., Gualdoni, S.M., Heilbrun, L.K., Valdivieso, M. and Bradley, E.C. (1990) Cancer 65, 2471. [18] Mitchell, M.S., Kempf, R.A., Harel, W., Shau, H., Boswell, W.D., Lind, S. and Bradley, E.C. (1988) J. Clin. Oncol. 6, 409. [19] Paciucci, P.A., Ryder, J.S. and Holland, J.F. (1990) Proc. Am. Soc. Clin. Oncol. 9, 201 (Abstr. 777). [20] Dutcher J.P., Gaynor, E.R., Boldt, D.H., Doroshow, J.H., Bar, M.H., Sznol, M., Mier, J., Sparano, J., Fisher, R.I., Weiss, G., Margolin, K., Aronson, F.R., Hawkins, M. and Atkins, M. (1991) J. Clin. Oncol. 9, 641. [21] McCabe, M.S., Stablein, D. and Hawkins, M.J. (1991) Proc. Am. Soc. Clin. Oncol. I0, 213 (Abstr. 714). [22] Mitchell, M.S., (1988) J. Natl. Cancer Inst. 80, 1445. [23] Kedar, E. Ben-Aziz, R., Epstein, E. and Leshem, B. (1989) Cancer lmmunol. Immunother. 29, 74. [24] Wolmark, N., Sicker, D., Yee, L., Hiserodt, J.C. and Salup, R.R. (1990) Proc. Am. Assoc. Cancer Res. 31,268 (Abstr. 1585). [25] Boldt, D.H., Mills, B.J., Gemlo, B.T., Holden, H., Meir, J., Paietta, E., McMannis, J.D., Escobedo, L.V., Sniecinski, I., Rayner, A.A., Hawkins, M.J., Atkins, M.B., Ciobanu, N. and Ellis, T.M. (1988) Cancer Res. 48, 4409. [26] Redman, B.G., Flaherty, L., Chou, T.H., Nakeff, A., Pillote, K. and Kaplan, J. (1991) J. Immunother. 10, 147. [27] Favrot, M.C., Combaret, V., Negrier, S., Philip, I., Thiesse, P., Freydel, C., Bijmann, J.T., Franks, C.R., Mercatello, A. and Philip, T. (1990) J. Biol. Response Modif. 9, 167. [28] West, W.H., Tauer, K.W., YanneUi, J.R., Marshall, G.D., Orr, D.W., Thurman, G.B. and Oldham, R.K. (1987) N. Engl. J. Med. 316, 898. [29] Blay, J.Y., Combaret, V., Negrier, S., Franks, C.R., Mercatello, A., Philip, T. and Favrot, M. (1991) Proc. Am. Soc. Clin. Oncol. 10, 217 (Abstr. 732). [30] Banerjee, D., Mertens, W., Bramwell, V. and Lala, P.K. (1991) Proc. Am. Assoc. Cancer Res. 32, 247 (Abstr. 1471). [31] Parmiani, G., Anichini, A. and Fossati, G. (1990) J. Natl. Cancer Inst. 82, 361. [32] Greenberg, P~D. (1991) Adv. lmmunot. 49, 281. [33] Parkinson, D.R. (1990) J. Natl. Cancer Inst. 82, 1374. [34] V~nky, F., Klein, E., Williems, J., B66k, K., Ivert, T., P&erffy, A., Nilsonne, U., Kreicbergs, A. and Aparisi, T. (1986) Cancer Immunol. Immunother. 21, 69. [35] Uchida, A., Kariya, Y., Okamoto, N., Sugie, K., Fujimoto, T. and Yagita, M. (1990) J. Natl. Cancer Inst. 82, 1697. [36] Ortaldo, J.R. and Wiltrout, R.H. (1990) J. Natl. Cancer Inst. 82, 1663. [37] Mitchell, M.S. (1991) Int. Rev. Immunol. 7, 331.