Low-Dose Interferon Gamma Renders Neuroblastoma Susceptible to Interleukin-2 Immunotherapy ByRobert

More

K. Sigal, Michael D. Lieberman, John V. Reynolds, Jian Shou, Moritz M. Ziegler, and John M. Daly Philadelphia, Pennsylvania

l Neuroblastoma remains a common and deadly childhood tumor, resistant to both surgical and chemo/radiotherapeutic intervention in its advanced stages. The role of immunotherapy in such cancers has yet to be defined. In previous work, we found that the addition of interferon gamma (IFN-r) to 3-day in vitro tissue cultures of the murine neuroblastoma C1300, led not only to the tumor’s increased cell surface expression of the immunologically important major histocompatibility complex (MHC) class I antigen, but also to an increased susceptibility of such modified tumor to subsequent lymphokine activated killer (LAK) cell lysis. In this study, we sought to determine the in vivo applicability of these findings. Initial dose-response studies helped define a regimen of rlFN-r’s administration that upregulated MHC class I without activating host natural killer (NK) activity. A/J mice bearing 7-day-old subcutaneous Cl300 were randomized to receive daily morning injections of either 0, 25,000, 50,000, or 100,000 U of rlFN-y intraperitoneally for 6 days. Animals were killed at days 3, 6, and g after initiation of rlFN-y therapy, and tumors were excised, digested, and stained for both MHC class I and II expression. At the time of sacrifice, splenocytes from each animal were tested for NK cytotoxicity toward YAC (an NK-sensitive lymphoma) and C1300. These studies defined 3 days of therapy with 25,000 U as a “priming” dose that increased expression of class I with minimal impact on NK activity. To test the effect of this dose of rlFN-y on subsequent LAK generation, splenocytes from three primed tumor-bearing animals were pooled and cultured in vitro with 1,000 U of recombinant interleukin-2 (rlL-2). Compared with controls, LAK cells from these primed animals were found to have an enhanced cytotoxicity toward both Cl300 and Cl300 modified in vitro by rlFN-r. When administered in vivo, sequential treatment with rlFN-y and rlL-2 (75,ooO U intraperitoneally, three times a day for 3 days) led to delays in tumor growth and significant (P < .05) increases in mean survival (46.7 days) when compared with rlL-2 (33.3 days), rlFN-r (31.7 days), or no treatment (28.1 days). These data help to define an immunotherapeutic regimen with which to treat human neuroblastoma. Copyright o 1991 by W.B. Saunders Company INDEX WORDS: Neuroblastoma, feron; interleukin-2.

immunotherapy;

inter-

N

EUROBLASTOMA (NB) remains a common and deadly childhood tumor. Although surgery alone can effect cure in 80% to 90% of the minority of children presenting with localized disease (Evans stage I or II),‘.’ regionally advanced or metastatic NB (stage III or IV) remains generally resistant to multimodal therapy.2-4 Despite a dismal prognosis for children with advanced disease, there exists a subpopulation of children with disseminated NB who have a decidedly good prognosis (stage IV-S). Spontaneous regression of the tumor from children with wideJournalofPeaiafric Surgery, Vol26, No 4 (April), 1991: pp 389-396

spread disease in the liver, skin, and/or bone marrow has been well documented.5.6 Although mechanisms for this regression remain ill-defined, the possibility of immunologically mediated rejection remains real. Thus, the comparatively new modality of immunotherapy may have particular impact on the treatment of NB. Interleukin-2 (IL-2) and interferon gamma (IFN-y) are biologic substances involved in the regulation of the immune system. The isolated genes coding for these lymphokines have been bacterially cloned, yielding large amounts of homogeneous recombinant IL-2 (rIL-2) and IFN-y (rIFN-Y).‘,’ Immunotherapy using high doses of rIL-2 alone or in combination with adoptive cellular therapy has mediated the regression of advanced malignancies in both animals’.l3 and humans.‘4-‘7 Similarly, rIFN-y therapy has resulted in the reduction of several murine tumors,‘8~‘9and has been studied in clinical protocols.2”-22 As with the development of chemotherapeutic treatments for cancer, in which single-agent therapies progressed to regimens involving multiple drugs, cytokines with individual therapeutic potential are now being studied in combination.‘3-25Like chemotherapeutic agents, characteristics inherent to each immunotherapeutic agent may suggest patterns of combination. For example, rIFN-y has been found to not only activate immune cells capable of killing tumor26.‘7and to directly inhibit tumor growth,28 but it also increases the amount of the immunologically important major histocompatibility complex (MHC) gene products expressed on tumor cell surfaces.“’ rIL-2 activated killer (LAK) cells have been found to preferentially lyse tumor cells with such increased levels of MHC gene product.30,31 From The Harrison Department of Surgical Research, The University of Pennsylvania School of Medicine, and The Joseph Stokes Research Institute, The Children’s Hospital of Philadelphia, Philadelphia, PA. Supported in part by the Georgene S. Harmelin Surgical Oncology Research Fund and by NC1 Grant No. 5-T32-CA 09619-o. Dr Sigal is a recipient of an NIH Cancer and Nutrition research fellowship. Presented at the 21st Annual Meeting of the American Pediatric Surgical Association, Vancouver, British Columbia, May 19-22, 1990. Address reprint requests to John M. Daly, MD. Chief Division of Surgical Oncology, Depatirnent of Surgery, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104. Copyright o 1991 by W.B. Saunders Company 0022-3468/91/2604-0006$03.OOlO 389

390

SIGAL ET AL

We have shown previously, that a murine NB can be induced to express more MHC class I gene product by the addition of rIFN-y to tissue culture.32 These modified cells were not only more immunogenic when injected back into animals,32,33but were more susceptible to in vitro LAK cell lysis than were unmodified control cells. In the current experiments, we sought to apply these findings to animals bearing the Cl300 NB subcutaneously. After defining a dosage regimen of rIFN-y that upregulated tumor MHC class I expression without activating natural killer (NK) cells, tumor-bearing animals were treated with rIFN-y followed by rIL-2. MATERIALS

AND METHODS

Mice All mice were male A/J (strain A) between 6 and 10 weeks of age. They were obtained from Jackson Laboratories (Bar Harbor, ME) and housed under climate-controlled, virus-free conditions, with regulated light/dark cycles.

Tumors The Cl300 NB originated spontaneously in strain A mice. Its histology as well as growth and immunizing characteristics have been previously described.j4”’ For the current experiments, an aliquot of frozen tumor was thawed and passaged in vivo. The resulting subcutaneous tumor was excised, mechanically dissociated, expanded in vitro (RPM1 1640 supplemented with 10% FCS, 1% L-glutamine, 100 ug/mL streptomycin, and 100 U/mL penicillin at 37°C in a 5% CO, incubator), then refrozen. When needed, aliquots were thawed from this stock, grown to a confluent monolayer, trypsinized briefly, and washed before use. Both the YAC-1 lymphoma (NK-sensitive) and the P815 mastocytoma (NK-resistant) were maintained in culture in this laboratory.

Subcutaneous Tumor Model Subcutaneous tumors were generated over the right flank of A/J mice by the subdennal injection of 1 x 10hlive Cl300 tumor cells in 0.05 mL of Hank’s phosphate buffered saline (PBS). Tumors achieved a diameter of 1.0 cm after 10 days. All size measurements were made by vernier caliper in two orthogonal directions, with tumor volume computed by the formula (P x length x width*/6).

Lymphokines Murine rIFN-y at 4 x lo6 U/mL was provided by Genentech (San Francisco, CA). For intraperitoneal (IP) injection, it was diluted to the appropriate concentration in phosphate buffered saline supplemented with 0.2% pooled autologous (A/J) serum as carrier. rIL-2 was provided by Hoffman LaRoche (Nutley, NJ) at a concentration of 2.8 x 10’ U/mL. It was similarly diluted for IP injection.

Antibodies Effects of the in vivo administration of rIFN-y on the expression of tumor cell MHC gene products were assessed by indirect immunofluorescence on a FACS IV autoanalyzer (Becton Dickinson, Mountain View, CA) using monoclonal antibodies from the following hybridomas (specificities): 16-l-1lN (H2K’), 34-1-2s (H2Kd, H2Dd), 26-7-118 (la”), and 17-3-38 (le”). Hybridomas were

procured from the American type culture collection and expanded to maximal cell density in the recommended media before the antibody-laden supernatant was cut with a 40% solution of ammonium sulfate for use as stain. Negative controls were stained with secondary antibody alone (fluoresceinated goat anti-mouse FAb,, Cappel, West Chester, PA), and splenocytes isolated from AKR mice served as positive controls.

Dose-Response Studies To define a “priming” dose of rIFN-y that would modify tumor without activating natural immunity, a series of dose-response experiments were performed. A/J mice (n = 12) bearing 7-day-old subcutaneous C1300, were randomized to four rIFN-y dosages: 0, 25,000,50,000, or 100,000 U, IP, every morning for 6 days. Animals were killed at days 3,6, and 9 after the initial rIFN-y injections. At the time of killing, tumors were excised, enzymatically digested (collagenase Type I 20 mg [Worthington, Freehold, NJ]; DNase 2 mg [Sigma, St Louis, MO], and hyaluronidase 100 U [Sigma, St Louis, MO] in 20 mL of RPM1 1640 with 25 mmol/L Hepes at 37°C for 1 hour), then stained with the above antibody panel. Concomitantly, each animal’s spleen was removed and assayed for NK activity by previously described techniques.” Briefly, single cell suspensions from each spleen were plated in 96 well conical bottomed plates with “Cr-labeled tumor targets (4 x lO’/well). In each case, triplicate wells were run at effector to target (E:T) ratios of lOO:l, 50:1, 25:1, and 12:l for cytotoxicity studies lasting 4 hours. Natural cytotoxicity was ascertained by measuring released gamma activity (Wallac Clinigamma, Gaithersburg, MD), and is expressed as % specific cytotoxicity ([sample counts - spontaneous counts]/[maximum counts - spontaneous counts]). To test the effects of a 3-day priming dose of 25,000 U of rIFN--r every morning on subsequent LAK cell generation, splenocytes from three primed tumor-bearing animals were pooled and cultured for 3 days in complete media with IL-2 added to a final concentration of 1,000 U/mL. LAK cells thus generated were compared with LAK cells similarly generated from unprimed tumor-bearing animals in their ability to lyse YAC-1, P815, C1300, and Cl300 modified by 3-day in vitro incubation in rIFN-y (C1300*, the in vitro equivalent of the in vivo model) in 4-hour ‘Cr release assays. This assay was repeated once with representative data reported.

Immunotherapy Model After defining a priming dose of rIFN-y for the model as 25,000 U, IP for 3 days, 32 mice with palpable (7-day-old) subcutaneous tumors were randomized to this dose or to daily injections of excipient buffer. After priming, each cohort was further randomized to receive 75,000 U rIL-2, IP, three times daily for 3 days or control injections. Animals were then followed for tumor growth and survival.

Statistics Determination of significance between treatment cohorts was made by ANOVA, with a Scheffe multiple comparison used to athx specificity. RESULTS

Different doses of a lymphokine with multiple effects emphasize different functions of that lymphokine. To define an in vivo dosage regimen of rIFN-y that upregulates tumor cell expression of MHC class I

IFN-y AND IL-2 IMMUNOTHERAPY

391

IN NEUROBLASTOMA

gene product and left natural immunity relatively unaffected, we initiated a dose-response experiment around doses previously demonstrated to affect MHC gene expression.39 Figure 1 is a series of composite fluorosceint-activated cell sorter (FACS) histograms obtained from stains of tumor digests of tumorbearing animals receiving 0,25,000,50,000, or 100,000 U of rIFN-y daily, IP, for 3 days. At the time of tumor excision, each was approximately 0.5 mL in volume (1 cm in diameter), with no areas grossly necrotic. These histograms demonstrate that although all doses of rIFN-y increase class I (H2Kk, H2Kd) expression, none significantly affect the cell surface expression of class II (Zuk, 1e”). Furthermore, whereas 25,000 U of rIFN-y increase mean class I expression by factors of 1.43 and 1.62 (H2Kk and H2Kd, respectively), treatment with both 50,000 and 100,000 U further augment that expression. Six-day treatment with these rIFN-y doses had no appreciable impact on class II, but had interesting effects on class I (H2Kk; Table 1). Whereas all doses of rIFN-y increased class I after 3 days of treatment, 6 days of treatment saw actual decreases in class I that varied directly with dose. That is, 25,000 U of rIFN-y caused no appreciable change in mean class I at 6 days (ratio = 1.01) but 50,000 U and 100,000 U produced factor decreases of 0.64 and 0.47, respectively. This pattern persisted 3 days after treatment end (day 9). Because we have not seen 6-day in vitro treatment affect Cl300 in this fashion3’ this selection for class I deficiency may be a function of in vivo rIFN-y administration. Possible explanations would rely on an activated immune surveillance associated with in vivo rIFN-y administration (Table 2) preferentially lysing cells expressing class I. Alternatively, in vivo administration of rIFN-y may have direct effects on tumor class I expression (perhaps associated with the focal areas of necrosis seen in large tumors) not appreciated in vitro. At the time of tumor excision, splenocytes in

1oD

IO’

102

IO3

IO’

IO2

l-lean Channel

Table 1. Class

I (H2K*)

Expressed by Cl300 After rlFN-y:

Dose Response Mean Channel Fluorescence (arbirtary

units)

Day 3

Day 6

Day 9

0

42.1

51.9

39.4

25,000

60.2

52.6

47.5

50,000

80.6

33.0

22.2

100,000

86.2

24.6

17.8

rlFNy Dose*

*Units IP every morning for 6 days.

treated animals were assayed for natural cytotoxicity to both YAC-1 and C1300. Figure 2A represents the cytotoxicity curves toward YAC-1 obtained from animals treated with 0,25,000,50,000, or 100,000 U of rIFN-y for 3 days. Although no cytotoxic effect is present at any E:T ratio for either 0 or 25,000 U treatments, both 50,000 and 100,000 U treatments are associated with aumented cytotoxicity at the higher E:T ratios. In Fig 2B, representing natural cytotoxicity toward the implanted Cl300 tumor, we note a small cytotoxicity in all treatment groups at the higher E:T ratios. However, this immunity is augmented by rIFN-y, with higher doses associated with more augmentation, Table 2 is a compilation of cytotoxicities toward both YAC-1 and Cl300 from days 6 and 9 after treatment initiation. Each value is from the 1OO:lE:T ratio. Toward YAC-1, we note a relative dissipation of NK activity as the tumor grows, an activity sustained by only the highest (100,000 U) regimen of rIFN-y. At the end of treatment, all regimens appear to have little impact on the development of immunity toward C1300, the tumor each animal was bearing. At day 6, all regimens, including the excipient control, manifest moderate levels of immunity. Three days after treatment end (day 9); however, only those animals treated with the largest amounts of rIFN-y continued to manifest immunity toward the tumor. These data suggested that 25,000 U of rIFN-y administered IP for 3 days augmented tumor MHC

IO3

IO’

IO2

1 o3

IO’

‘02

Fluorescence

Fig 1. Composite FACS histograms demonstrating class I (H2K‘.H2P’) and class II (la’, /e*) expressed by C1300. Ten-day-old tumors were excised and digested after 3-day host therapy with rlFN-y or excipient buffer. Note dose-related increased expression of class I, with minimal effect on class II.

SIGAL ETAL

392

Table 2. NK Activity After rlFN-y: Dose Response Cytotoxicity

(%)t

v YAC-1 rlFN--, Dose*

v Cl300

Day 6

Day 9

1

0

16.3

0

25,000

0

0

18.8

0

50,000

3.5

0

13.4

0

100,000

18.8

0

18.9

7.8

0

*Units

IPevery morning

for

Day 6

Day 9

6 days.

tE:Tratio= 1OO:l.

expression with minimal impact on the recipient’s natural immunity. However, before moving to in vivo administration, we sought to determine the effect of this regimen on the generation of LAK cells. Figure 3 is an amalgam of cytotoxicity curves obtained when splenocytes from rIFN-y-primed tumor-bearing animals were subsequently cultured in vitro with rIL-2. Against YAC-1 and P815, LAK cells from rIFN-yprimed animals had similar cytotoxicity curves to excipient controls. However, against Cl300 and C1300*, rIFN-y treatment enhanced LAK cytotoxicity. Figures 4 and 5 represent tumor growth and survival curves from the four treated cohorts. In Fig 4, tumor growth assumes similar patterns in each treatment group; however, the addition of each lymphokine adds a delay (shift to the right) to the inexorable growth of the tumor. Tumor measurements cease after the first death in a cohort. This first occurred in animals treated only with buffer, then treated only with rIFN-y, then with rIL-2, and finally with both lymphokines. This pattern was reflected in mean survival (Fig 5); animals treated with buffer lived on average 28.1 days, animals treated with rIFN-y lived 31.7 days, rIL-2 lived 33.3 days, and animals treated first with rIFN-y then rIL-2 lived 48.7 days (P < .05). DISCUSSION

Interest in the use of immunotherapy as a cancer treatment has been stimulated by published reports YAC30-

Fig 3. LAK cytotoxicity toward tumor targets after treatment with excipient buffer or 25,000 U of rlFN-7, IP for 3 days (rlFN-7). Splenocytes from treated tumor-bearing animals were cultured for 3 days with 1,000 U/ml_ rlL-2 before standard 4-hour “Cr release assays (*). Data given as mean of triplicate wells.

demonstrating the regression of metastatic cancer in selected patients by the administration of rIL-2 alone or in conjunction with LAK or tumor infiltrating (TIL) cells. This potentially powerful addition to the surgeon’s armamentarium has been hampered by its toxicity and ineffectiveness against the more prevalent solid tumors. One approach toward solving both of these problems is to combine a number of cytokines, each with different mechanisms of action. The combination of rIL-2 and rIFN-y is one combination of particular interest. Detailed knowledge of the mechanism of each lymphokine’s antitumor effects will aid in designing combination therapies. rIL-2 mediates its therapeutic effects by both the in vivo generation of LAK cells, and by the stimulation of Lyt-2’ T cells with a more selective and more powerful antitumor reactivity.40,4’ These latter cells appear to be especially important to the effective therapy of tumor in that their elimination abrogates otherwise successful rIL-2-based regimens.42 Because Cl300

1 _

A

B

rlfnv dose: -0 25,000~ --+-50,000u

.

-

.

100,000u

Bej : , &* 1OO:l

5O:l E:T ratio

Fig 2. Natural cytotoxicity toward (A) YAC-1 and (B) C1300. After 3 days of rlFN-7, splenocYtes from animals bearing loday-old Cl300 manifest cytotoxicity toward YAC-1 only with high

1 ST 25:l

12:l

1OO:l

~ , ~~~~~~~~

5O:l E:l

25:l ratio

12:l

with more cytocidal activity. Data given as mean of triplicate wells.

IFN-7 AND IL-2 IMMUNOTHERAPY

0

393

IN NEUROSLASTOMA

20

10 Day after

30

40

hoculatian

Fig 4. Tumor growth versus time for cohorts of animals treated with elther excipient buffer, 25,WO U rlFN-y IP a day for 3 days, 75,ooO U rlL-2 IP three times a day for 3 days, or both lymphokines sequentially. Tumor measurements represent mean (ISEM) of calculated tumor volumes and cease after the first death in a cohort.

these cells function through the T cell receptor and antigens are presented to them in the context of MHC class I gene products, the expression of class I on the surface of target cells is vital to their function. In demonstration of this, Weber et a131have shown an increased sensitivity to rIL-2 therapy after transfection of MHC class I genes into tumor cells otherwise devoid of expressed class I. Interferons, although structurally related, are a heterogeneous group of biologically active molecules produced by different cells. All have the ability to stimulate natural immunity, hinder (tumor) cell growth, and upregulate the cellular expression of MHC class I and II gene products, but each to varying degree. For example, of the interferons, interferon alfa (rIFN-ol), also known as leukocyte interferon, appears to have the greatest impact on NK activation. Therefore, it has been the focus of the most clinical interest. High doses have been used to treat patients with leukemia,43,44renal cell cancer,45 and melanoma.& Concomitant (as opposed to sequential) administration of rIFN-ol and rIL-2 has been the rule in animals studies,23,24and has proven more effective than either lymphokine alone in phase I clinical trials:’ In contrast to rIFN-a, rIFN-y, or lymphocyte interferon, has less NK activating potential, and more effect on the MHC gene locus. It too has been used and found to be additive when administered concomitantly with and clinical trials.49 However, rIL-2 in both animals25948 because of rIFN-y’s impact on the MHC class I locus, we felt sequential (rather than concomitant) therapy with rIL-2 might prove a more rational approach to their combined administration.

The effect of rIFN-7 on tumor class I expression may have particular relevance for NB, a tumor notoriously low in expressed class 1.5’Lampson et al have found such NB cells to be relatively insensitive to T-cell-mediated lysis.” Attempting to modify class I expression, Evans et al recently treated 7 children with advanced NB with low-dose IFN-y and followed the class I expression by tumor in bone marrow, as well as NK and LAK precursor activity in the blood?’ Although class I was initially absent on the neuroblastoma cells, rIFN-y induced HLA-A,B,C in four of six evaluable patients. The invasive nature of the bone marrow biopsy precluded repetitive monitoring of the expressed class I. NK activity, initially low, transiently returned to baseline, whereas LAK precursor activity remained normal. In the current study, rIFN-y was also capable of increasing the class I expressed on NB cells when administered in vivo, although this effect appeared to be transient. Importantly, following this induction, IL-2 administration proved more efficacious in controlling tumor growth. The in vitro analyses accompanying the rIFN-y priming dose that examined subsequent LAK generation (Fig 3) suggest that this effect was not solely due to target cell modification. Although NK cell activity was unaffected by rIFN-y at 25,000 U (Fig 2) LAK precursor activity toward Cl300 and C1300* was increased in primed tumorbearing animals. Recent work by Ellis et al offers partial explanation in that IFN-y was found to play a participating role in the optimal induction of LAK cells by IL-2.52 The suggestion for these studies arose from in vitro work that found NB cells expressing increased levels of MHC class I more susceptible to LAK cell lysis.

100 no tx/no tx

l---l

80 5 ‘C ?

___---.._.

rlfny/no tx ----no tx/r IL-2 ..... ....... ... rlfny/rfL_2

60

*____, I_, 0,.

.

0

.

13,.

\

;

.I.

*,

40

20 Day

:

:

after

i

*: 60

-

I

80

lnnoculrtion

Fig 5. Host survival over time for cohorts treated with either buffer, r1FN-y. rll_-2. or both lymphokinas sequentially. Animals treated with both rlFN-r and rlL-2 demonstrated prolonged survival compared with all other treatment groups (P i .05).

394

SIGAL ET AL

Our finding in this study that primed cytotoxicity toward C1300* was not appreciably higher than against Cl300 (Fig 3) runs counter to previous findings using unprimed LAK cells. Possible explanation may lie in the priming process itself, and its recruitment of LAK cell precursors toward a less class

I-responsive phenotype. This would suggest that future studies focus on optimal regimens for effector cell generation rather than on tumor modification. Nevertheless, the current study provides impetus for clinical trials of sequential immunotherapy using IFN-?/ and rIL-2 in advanced NB.

REFERENCES 1. Nitschke R, Smith EI, Shochat S, et al: Localized neuroblastoma treated by surgery: A pediatric oncology group study. J Clin Oncol6:1271-1279,1988 2. Evans AE, D’Angio GJ, Koop CE: The role of multimodal therapy in patients with local and regional neuroblastoma. J Pediatr Surg 19:77-80, 1984 3. Berthold F, Brandeis WE, Lanpert F: Neuroblastoma: Diagnostic advances and therapeutic results in 370 patients. Monogr Paediatr 18:206-223,1986 4. Voute PA, Vos A, Delemarre JFM, et al: The persistent challenge of neuroblastoma, in van Eys J, Sullivan MP (eds): Status of the Curability of Childhood Cancer. New York, NY, Raven, 1980, pp 145161 5. D’Angio GJ, Evans AE, Koop CE: Special pattern of widespread neuroblastoma with a favourable prognosis. Lancet 1:10461049,197l 6. Nickerson HJ, Nesbit ME, Grosfeld JL, et al: Comparison of stage IV and IV-S neuroblastoma in the first year of life. Med Pediatr Oncol 13:261-268,1985 7. Rosenberg SA, Grimm EA, McGrogan M, et al: Biological activity of recombinant human interleukin-2 produced in E coli. Science 223:1412-14151984 8. Houghton AN, Thomas TM, Gross D, et al: Surface antigens of melanoma and melanocytes. Specificity of induction of la antigens by human IFN. J Exp Med 160:255-269,1984 9. Rosenberg SA, Mule JJ, Speiss PJ, et al: Regression of established pulmonary metastases and subcutaneous tumor mediated by the systemic administration of high dose recombinant interleukin-2. J Exp Med 161:1169-1188,1985 10. Mule JJ, Shu S, Schwarz SL, et al: Successful adoptive immunotherapy of established pulmonary metastases with LAK cells and recombinant rIL-2. Science 225:1487-1489, 1986 11. Lafreniere R, Rosenberg SA: Adoptive immunotherapy of experimental hepatic metastases with lymphokine activated killer cells (LAK) and recombinant interleukin-2 can mediate the regression of both immunogenic and nonimmunogenic sarcomas and an adenocarcinoma in a murine system. J Immunol 135:4273-4280, 1985 12. Speiss PJ, Yang JC, Rosenberg SA: Tumor infiltrating lymphocytes expanded in recombinant interleukin-2 mediate potent antitumor activity in vivo. JNCI 79:1067-1075, 1988 13. Rosenberg SA, Speiss PJ, Lafreniere R. A new approach to the adoptive immunotherapy of cancer with tumor infiltrating lymphocytes. Science 233:1318-1321,1986 14. Rosenberg SA, Lotze MT, Muul LM, et al: Observations on the systemic administration of autologous lymphokine-activated killer cells and recombinant interleukin-2 to patients with metastatic cancer. N Engl J Med 313:1485-1492,1985 15. Rosenberg SA, Lotze MT, Muul LM, et al: A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and interleukin-2 or high dose interleukin-2 alone. N Engl J Med 316:889-897, 1987 16. West WH, Tauer KW, Tannelli JR, et al: Constant-infusion recombinant interleukin-2 in adoptive immunotherapy of advanced cancer. N Engl J Med 316:898-905,1987 17. Topalian SL, Solomon D, Avis FP, et al: Immunotherapy of

patients with advanced cancer using tumor-infiltrating lymphocytes and recombinant interleukin-2: A pilot study. J Clin Oncol 6:839853,1988 18. Jaffe HS, Herberman RB: Tationale for recombinant human interferon-gamma adjuvant immunotherapy for cancer. JNCI 80: 616-618, 1988 19. Ijzermans JN, Marquet RL, Bouman E, et al: Successful treatment of colon cancer in rats with recombinant interferongamma. Br J Cancer 56:795-796,1987 20. Brown TD, Koeller J, Beougher K, et al: A phase I clinical trial of recombinant DNA gamma interferon. J Clin Oncol 5:790798,1987 21. Ernstoff MS, Trautman T, Davis CA, et al: A randomized phase I/II study of continuous versus intermittent intravenous interferon-gamma in patients with metastatic melanoma. J Clin Oncol5:1804-1810,1987 22. Kaplan EH, Rosen ST, Norris DB, et al: Phase II study of recombinant human interferon gamma for treatment of cutaneous T-cell lymphoma. JNCI 82:208-212,199O 23. McIntosh JK, Mule JJ, Krosnick JA, et al: Combination cytokine immunotherapy with tumor necrosis factor alpha, interleukin-2, and alpha interferon and its synergistic antitumor effects in mice. Cancer Res 49:1408-1414,1989 24. Iigo M, Sakurai M, Tamura T, et al: In vivo antitumor activity of multiple injections of recombinant interleukin 2, alone and in combination with three different types of recombinant interferon, on various sungeneic murine tumors. Cancer Res 48:260-264,1988 25. Agah R, Malloy B, Sherrod A, et al: Successful therapy of natural killer-resistant pulmonary metastases by the synergism of gamma interferon with tumor necrosis factor and interleukin-2 in mice. Cancer Res 48:2245-2248,1988 26. Pace JL, Russell SW, Torres BA, et al: Recombinant mouse gamma interferon induces the priming step in macrophage activation for tumor cell killing. J Immunol 130:2011-2013, 1983 27. Weiner LM, Moldofsky PJ, Gatenby RA, et al: Antibody delivery and effector cell activation in a phase II trial of recombinant gamma interferon aan the murine monoclonal antibody C017-1A in advanced colorectal carcinoma. Cancer Res 48:25682573,1988 28. Pfizenmaier K, Bartsch H, Scheurich P, et al: Differential gamma-interferon response of human colon carcinoma cells: Inhibition of proliferation and modulation of immunogenicity as independent effects of gamma-interferon on tumor cell growth. Cancer Res 45:3503-3509,1985 29. Skoskiewicz MJ, Colvin RB, Schneeberger EE, et al: Widespread and selective induction of major histocompatibility complexdetermined antigens in vivo by gamma interferon. J Exp Med 162:1645-16641985 30. Greiner JW, Guadagni F, Noguchi P, et al: Recombinant interferon enhances monoclonal antibody targeting of carcinoma lesions in vivo. Science 235:895-898, 1987 31. Weber JS, Jay G, Tanaka K, et al: Immunotherapy of a murine tumor with interleukin 2: Increased sensitivity after MHC Class I gene transfection. J Exp Med 166:1716-1733,1987 32. Sigal RK, Reynolds JV, Markmann JF, et al: Upregulation

IFN-?, AND IL-2 IMMUNOTHERAPY

395

IN NEUROBLASTOMA

of MHC class I: Effect on growth and LAK sensitivity of neuroblastoma. Surg Forum 39:572-575,198s 33. Sigal RK, Reynolds JV, Lieberman ML, et al: Tumor immunization: Improved results after recombinant gamma interferon modification. Arch Surg 125:308-312, 1990 34. Ziegler MM, Kirby J, McCarrick JW III, et al: Neuroblastoma and nutritional support: Influence on the host-tumor relationship. J Pediatr Surg 21:236-239,1986 35. McCarrick JW III, Ikeda CB, Ziegler MM: Tumor immunogenicity-The prime determinant of the nutritional influence on the host-tumor relationship. J Parenter Enter Nutr 10:21-28, 1986 36. McAlack RF, Buck BE, Hicks N, et al: Metastatic characteristics of murine neuroblastoma, C1300. Fed Proc 26:1086,1977 37. Ziegler MM, Naito H, McCarrick JW. et al: Cl300 murine neuroblastoma: A suitable animal model of human disease, in Brooks BF (ed): Malignant Tumors of Childhood. Austin, TX, University of Texas, 1986, pp 114-126 38. Reynolds JV, Shou J, Sigal RK, et al: The influence of natural killer cells in neuroblastoma. Arch Surg 124:235-239, 1989 39. Skoskiewicz MJ, Colvin RB, Schneeberger EE, et al: Widespread and selective induction of major histocompatibility complexdetermined antigens in vivo by gamma interferon. J Exp Med 162:1645-1664,1985 40. Mule JJ, Shu S, Schwarz SL, et al: Adoptive immunotherapy of established pulmonary metastases with LAK cells and recombinant interleukin-2. Science 225:1487-1489, 1984 41. Rosenberg SA, Mule JJ, Spiess PJ, et al: Regression of established pulmonary metastases with lymphokine-activated killer cells and recombinant interleukin-2. J Exp Med 161:1169-1188, 1985 42. Mule JJ, Yang JC, Lafreniere R, et al: Identification of cellular mechanisms operational in vivo during the regression of established pulmonary metastases by the systemic administration

of high-dose recombinant interleukin-2. J Immunol 139:285-294, 1987 43. Talpaz M, Katarjian HM, McCredie K, et al: Hematologic remission and cytogenetic improvement induced by recombinant human interferon-alpha in chronic myelogenous leukemia. N Engl J Med 314:1065-1069,1986 44. Foon KA, Bottino GC, Abrams PG: Phase II trial of recombinant leukocyte A interferon in patients with advanced chronic lymphocytic leukemia. Am J Med 78:216-220, 1985 45. Muss HB: Interferon therapy for renal cell carcinoma. Semin Oncol 14:36-42, 1987 (suppl2) 46. Robinson WA, Mughal TI, Thomas MR, et al: Treatment of metastatic melanoma with recombinant interferon-alpha 2. Immunobiology 172:275-282, 1986 47. Rosenberg SA, Schwarz SL, Spiess PJ: Combination immunotherapyfor cancer: Synergistic antitumor interactions of interleukin-2, alfa interferon, and tumor-infiltrating lymphocytes. JNCI S&1393-1397, 1988 48. Silagi S, Dutkowski, Schaefer A: Eradication of mouse melanoma by combined treatment with recombinant human interleukin 2 and recombinant murine interferon-gamma. Int J Cancer 41:315-322,198s 49. Wagstaff J, Vermorken JB, Schwartsmann G, et al: A progress report of a phase I study of interferon-gamma and interleukin-2 and some comments on the mechanism of the toxicity due to interleukin-2. Cancer Treat Rev 16:105-109, 1989 (suppl A) 50. Lampson LA, Fisher CA, Whelan JP: Striking paucity of HLA-A,B,C and B2 microglobulin on human neuroblastoma cell line. J Immunol 130:2471-2478, 1983 51. Evans A, Main E, Zier K, et al: The effects of gamma interferon on the natural killer and tumor cells of children with neuroblastoma. Cancer 64:1383-1387.1989 52. Ellis TM, McKenzie RS, Simms PE, et al: Induction of human lymphokine-activated killer cells by IFN-alpha and IFNgamma. J Immunol 143:4282-4286.1989

Discussion G. Huase (Denver, CO): The problem clinically in NB of the advanced or resistant type is of course not the local tumor but the metastatic disease, and I am wondering if certainly in your study and perhaps in the previous one, if there is any hint that you’ve had any experience with using the biologic response modifiers in a postoperative mode. In other words, in a metastasizing system where the primary tumor is resected and then using your modifiers postoperatively to see if in fact you can enhance survival in the in vivo model. As I looked at your Kaplan-Meier curve, the one that you showed, I was not that impressed that the area under the curve differences were that great in the middle part or the middle range of the curve where most of the observations are made. It isn’t until you get out a long distance where perhaps there are less observations that you start to see a separation of the curves, and I think from the clinical stand point that would be most relevant. If you haven’t done that previously, you might consider that for future studies. This is a very nice laboratory study.

R.K. Sigal (response): I think that’s a very good point. Surgery I think is going to remain a mainstay of treatment in potentially curable NB and immunotherapy may well find a niche as an adjuvant in that setting. There have been a number of studies in animals that have looked at the ability of cells to metastasize once they have been modified by various mechanisms to change their MHC class expression and have found that up regulated or augmented MHCs were associated with decreased metastatic ability. So it may well be that immunotherapy finds an adjuvant role in potentially curable NBS. S.J. Shochat (Stanford, CA): I think this is a very significant study and in fact the principles of this particular study are going to be followed in a clinical study at NIH in which patients with advanced NB will be treated with interferon followed by interleukin-2 and tumor infiltrating lymphocytes as a phase II study that I think will be initiated in the next 6 to 12 months. R. K. Sigul (response): Although IFN-y upregulates class I both in vitro and in vivo in the Cl300 tumor, I

SIGAL ET AL

think we were very careful in this paper not to specifically ascribe our effects to that mechanism alone. Specifically, what we wanted to bring out in this paper besides the upregulation of MHC was the fact that this low dose of IFN-y appeared to prime LAK cell precursors toward the implanted tumor. Now that in itself may be due to a MHC upregulation and that perhaps a tumor with upregulated MHC is more visible to the immune system and allows a more effective cytotoxic T cell generation. There have been studies in humans in which patients were administered interferon and then NB cells were aspirated

from their bone marrow and assayed for class I. These studies were done by Dr Andrey Evans and the results were published just last year. In 4 of 6 evaluable patients, they found that inferon did in fact upregulate class I on the tumor cells that they aspirated. It may be that MHC is not critical to successful immunotherapy in NB. However, it may serve as a marker for other cell surface antigens that get upregulated and in that case it would be important to follow in that we don’t know what those antigens are, but we can follow MHC class I.

Low-dose interferon gamma renders neuroblastoma more susceptible to interleukin-2 immunotherapy.

Neuroblastoma remains a common and deadly childhood tumor, resistant to both surgical and chemo/radiotherapeutic intervention in its advanced stages. ...
945KB Sizes 0 Downloads 0 Views