Cytokine induction by 41.8 °C whole body hyperthermia

CancerLetters 97 (1995) 1’35-201

ELSEVIER

CANCER LETTERS

Cytokine induction by 41.8”C whole body hyperthermia H. Ian Robins+*, Matthew Kutza, Gunther J. Wiedemannb, D&the M. Katschinskib, Devchand Paula, Elizebeth Grosena, Cynthia L. Tiggelaara, David Spriggsc, Wilma Gillisa, Floriane d’ Oleirea of Wisconsin Clinical Science Center: Madison, WI, USA bMedical University of Ltibeck, Ltibeck, Germany ‘Memorial Sloan Kettering Cancer Center New York, USA

aUniversity

Received10August 1995;accepted24 August 1995

Abstract

The potential for 41.8”C whole body hyperthermia (WBH) to enhance ionizing irradiation and cytotoxic chemotherapy without a commensurate increase in normal tissue toxicity is currently receiving renewed clinical interest. Additionally, WBH may have other biological sequela which may be clinically exploited. In this paper, data are summarized revealing the ability of WBH to induce elevated plasma levels of granulocyte-colony stimulating factor (G-CSF), interleukin-lp (IL-l/l), interleukin-6 (IL-6), interleukin-8 (IL-S), interleukin-10 (IL-lo), and tumor necrosis factor-a (TNF-a) within hours of WBH. Data regarding TNF-a shows induction in only a proportion of patients. No induction of C-reactive protein (CRP) or the following cytokines was observed: granulocyte macrophage-colony stimulating factor (GM-CSF), interferon-y (IFN-y), interleukin-la (IL-la), interleukin-2 (IL2), interleukin-4 (IL-4), interleukin-7 (IL-7). interleukin-11 (IL-ll), interleukin-12 (IL-l 2), macrophage-colony stimulating factor (M-CSF), and macrophage inflammatory protein-la (MIP-la). Data regarding interleukin-3 (IL-3) and transforming growth factor-B1 (TGF-#Il) were variable and inconclusive. The implications of these results to past and future clinical trials are discussed. Keywords:

Cytokine induction; Whole body hyperthermia

1. Introduction Whole body hyperthermia (WBH) is undergoing renewed interest as an adjunct to cytotoxic therapy for the treatment of neoplastic disease. In this regard, an international cooperative group has been formed to share technology and protocols [7]. Clinical data [ 8,11,19] supports preclinical observations that WBH * Corresponding Highland Avenue,

author. Department of Medicine Madison, WI 53792-6164, USA.

0304-3835/95/$09.50 0 1995 Elsevier SSDI 0304-3835(95)03976-4

Science

K4/666,

600

can enhance cytotoxic ionizing irradiation and chemotherapy without a commensurate increase in myelosuppression. An explanation for these observations may reside in the induction of cytokines by 41.8”C WBH. A preliminary communication [l] demonstrated the induction of granulocyte colony stimulating factor (G-CSF) and interleukin (IL)-1s; tumor necrosis factor (TNF)-a was found to be induced in some but not all WBH patients. In the course of this study we assayed the plasma of WBH patients for induction of: G-CSF; granulocyte macro-

Ireland Ltd. All rights reserved

H.I. Robins et al. I Cancer

196

phage-colony stimulating factor (GM-CSF); interferon (IFN)-y; IL-la; IL-2; IL-3; IL-4; IL-6; IL-7; IL-8; IL- 10; IL- 11; IL- 12; macrophage-colony stimulating factor (M-CSF), and macrophage inflammatory protein- la (MIP- la); transforming growth factor-/31 (TGFj31); TNF-a; C reactive protein (CRP). Our collective results regarding these assays are presented; the implications of cytokine induction in the setting of WBH are discussed.

2. Materials

and methods

In our previous cytokine evaluations [l], some plasma samples were freeze thawed up to three times prior to assay. For our current report, however, samples were only thawed once at the time of the assay. In addition to assaying samples collected from radiant-heat WBH, selected samples collected from extra-corporeal WBH were also evaluated. Plasma was obtained before (15 min), during (at peak temperature), and after (end of peak temperature, 2.5 h, 5.5 h, and 23.5 h post) 60 min of 41.8”C whole body hyperthermia. For control data, plasma samples were also obtained from selected WBH patients who received chemotherapy alone. Samples were drawn in exactly the same time sequence as above (with chemotherapy substituting for 4 1.8”C WBH). Plasma was collected using EDTA or heparin as anticoagulants; in kits where heparin was known to interfere (IL-3, IL-4, TGF-/?l), EDTA collected samples were used. Samples were acquired from both the University of Wisconsin Clinical Sciences Center (UWCSC), USA, which utilizes radiant heat WBH (i.e. The Aquatherm) [6] and the University of Liibeck, Germany, which utilizes extra-corporeal WBH [19]. The time-temperature profiles of the two WBH methodologies are comparable (i.e. time to temperature; time at peak temperature; 41.8”C x 60 min; cooling). At the time of collection, plasma was aliquoted in multiple vials so that samples were frozen (-70°C) and thawed only at the time of the assay. Frozen samples were transferred from Lilbeck to the UWCSC on dry ice. Informed consent was obtained from all patients involved in these institutional review board approved studies. Cytokines were assayed using the following quantitative, commercially available enzyme linked immunosorbent assay (ELISA) kits: G-CSF, GM-

Letters

97 (1995)

195-201

CSF, IL-la, IL-2, IL-3, IL-4, IL-6, IL-7, IL-8, IL-l 1, M-CSF, MIP-la, TNF-a, and TGF-PI (R&D Systems, Inc. Minneapolis, MN); IFN-y and IL-10 (Genzyme Corp., Cambridge, MA). The following are minimum detectable levels using these assays: 10.9 pg/ml G-CSF; 2.8 pg/ml GM-CSF; 100 pg/ml IFN-y; 0.2 pg/ml IL-la; 10 pg/ml IL-2; 7.4 pg/ml IL3; 4.1 pglml IL-4; 0.7 pg/ml IL-6; 6 pg/ml IL-7; 18.1 pg/ml IL-8; 5 pg/ml IL-lo; 4pg/ml IL-11; 20 pg/ml M-CSF; 3 pg/ml MIP-la; 4.4 pg/ml TNFa; 5 pg/ml TGF-/Il. Samples that showed induction of IL-6 were assayed for the induction of CRP by the UWCSC clinical laboratories using the Abbot TDX fluorescent polarization immunoassay. The minimum detectable level of CRP with this method is 20 pg/ml.

3. Results 3.1. Radiant heat WBH Fig. 1 demonstrates the induction of IL-6. One patient in this assay series received chemotherapy alone, and was assayed for IL-6 at the comparable time points; there was no detectable IL-6 levels preand post-chemotherapy. Levels of IL-6 in normal patients for the assay kit ranged from 0 to 12.5 pg/ml. The current data reported on IL-6 differ from our previous report in which no induction of IL-6 was found in samples that had been freeze thawed repeatedly. Fig. 1 also demonstrates the induction of IL-8. Two patients in this series received chemotherapy alone for the comparable six time points. These two patients had constitutive levels of IL-8 of 38.7 f 10.6 pg/ml, but no increases post chemotherapy. The last graph in Fig. 1 demonstrates the induction of IL-lo. Two patients in this series were assayed for IL-10 after receiving chemotherapy alone for the comparable six time points. Both patients had no detectable levels of IL- 10. Fig. 2 shows the induction of G-CSF. These results are comparable with those that we have previously reported [ 11. Fig. 2 shows the induction of TNF-a by WBH for three of seven patients. Four other patients were assayed; three of these had no detectable level of TNFa pre- and post-WBH. One of these patients had a constitutive level of TNF-a of 9.4 r 5.8 pg/ ml.

H.I. Robins et al. I Cancer

Letters

97 (1995)

19.5-201

197

400 * 350 I! 300 g

250

a

200

9

150 100 50 0 4; .8’C

120

7

1 100

h

T

0

2

80

F

60

2 z

40

A 20 En

Cl

/

/

3

(GEM)

I. Radiant

heat

for n number

WBH of WBH

induction

of cytokines

treatments.

1

(hours

post

6 Time

Fig.

”

(IL-6,

The thatched

c

9

IL-&

IL-IO)

area represents

Five patients were assayed pre- and post-WBH for CRP. No significant changes were observed. An assay of seven patients for GM-CSF demon-

!

7

’

1 ;! 41.8%

“I’

t

15

18

I

“I

21

24

lhyperthertnia)

in plasm.1 time

I

samples.

(60 min)

at peak

Squares

(0)

temperature

represent

time

point

mean

values

(41.8”C).

strated no detectable plasma levels pre- and postWBH. This is consistent with our previous report [ 11. An assay of seven patients for IFN-y demonstrated

198

H.I. Robins et al. I Cancer

1 -3

0

3

6 Time (hours

I

Letters

I1

97 (199s)

195-201

I,

9 12 15 post 41 .VC hyperlhermia)

I

16

I

21

1

I

24

Fig. 2. Radiant heat WBH induction of cytokines (G-CSF and TNF-a) in plasma samples. Circles (0) represent time point mean values (GEM) for n number of WBH treatments. The thatched area represents time (60 min) at peak temperature (41.8”C). TNF data reflects the three out of seven treatments that showed detectable induction; four treatments were censored as there was no detectable TNF induction.

no detectable plasma levels pre- and post-WBH. This is consistent with the previous report [ 11. We found no detectable levels of IL-la, IL-7, IL12, M-CSF, and MIP- la for 7 patients assayed preand post-WBH. Similarly we found no detectable levels for IL-2 (as previously reported [I] in a different series of patients) and IL- 11 in 6 patients assayed. IL-3 is known to have a half-life of only 40 min [ 181. Fluctuation of IL-3 levels with both radiant heat and extra-corporeal WBH (three and two treatments, respectively) as well as one control (i.e. chemotherapy alone) were inconsistent and not statistically valid making any assessment for this cytokine impossible.

3.2. Extra-corporeal

WBH

GM-CSF was assayed in Germany after extracorporeal WBH (n = 6) with no detectable levels being found pre- and post-WBH. Comparable induction of G-CSF (n = 6) as shown above was found [19]. Extra-corporeal WBH samples assayed for IL-2 (n = 3), IL-3 (n = 2), IL-4 (n = 2), and IL-10 (n = 1) yielded results that were qualitatively and quantitatively comparable to those reported above by radiant heat WB’H. The results of pilot data regarding increased plasma llevels of TGF-@l are shown for both extra-

H.I. Robins et al. I Cancer

0 -3

0

3

6 Time

Fig. 3. Active TGF-PI plasma levels are shown as a function treatments (&EM) with radiant heat WBH. The open circles area

represents

time

(60 min)

at peak

temperature

(hours

Letters

97 (1995)

195-201

I 12 15 9 post 41 .VC hyperthermia)

”

199

I 18

”

I 21

“I

of time; i.e. pre- and post-WBH. The closed circles (0) (0) represent two patient treatments with extra corporeal

24

represent WBH.

three patient The thatched

(41.8”C).

corporeal and radiant heat WBH (see Fig. 3). Modest elevation for extra-corporeal WBH was noted. No changes, however, in plasma levels for radiant heat WBH were observed. 4. Discussion

WBH, but not IL-la, is consistent with the experience in patients with sepsis and fever [2] (in spite of the greater assay sensitivity for the IL-la assay). The lack of IL-la induction is also in accord with the observation that tissue culture conditions that cause the release of IL-16 do not cause the release of IL-la

PI. Hyperthermia enhances neoplastic cell kill by various genotoxic therapies, but has variable effects on hematopoietic tissue [ 121. Under specified conditions, radiant heat WBH was combined with cytotoxic chemotherapy or ionizing irradiation in both murine systems [11,16] and clinical trials [8,11]. Analysis of data from these investigations suggested the possibility of myeloprotective aspects of hyperthermia. Beyond this, Robins et al. [lo] reported early implantation of transplanted allogenic bone marrow when WBH was a component of pretransplant conditioning regimens. These observations are consistent with the induction of cytokines summarized in this report Clearly the induction G-CSF in combination with IL-l/3 (reported previously by our group [ 11) as well as IL-6 may have significant implications for the stimulation of myeloid, erythroid, and megakaryocytic progenitor cells [2,5,13,18]. Hence, the documented increases in the levels of these factors partially explain the aforementioned observations. It is of interest to note that the induction of IL-l@ by

It would be predictable from the above discussion that WBH might increase both platelet and white blood cell (WBC) counts in patients based on cytokine induction. The interpretation of data regarding WBCs is confounded by the fact that WBH induces the release of adrenocorticotropic hormone (ACTH) and adrenocorticoids [9,12], which affects WBC margination. Hormonal effects, however, are not obviously relevant to platelet counts. Unfortunately, human patient WBH treatment data are not interpretable by virtue of sequential treatments and/or multimodality therapy directed at treating underlying malignancies. We, therefore, tested the hypothesis that WBH would induce increased platelet counts in dogs. Results demonstrated a statistically significant increase in platelet counts day 9 to day 16 post- WBH [20]. This was also observed in mice day 8 postWBH [ll]. It is an interesting point of conjecture that the major endocrine changes that have been observed during WBH (i.e. endorphin, ACTH, cortisol, and prolactin elevations [12]) may be mediated by cy-

200

H.I. Robins et al. / Cancer

tokine induction. These would include IL-l stimulation of endorphin, ACTH [2,4], and cortisol release, and IL-6 induction of prolactin [ 143. As was previously reported [l], we found only a portion of patients with significant increases in TNFa. This may relate to the sensitivity of the assay, or significant biological differences between patients. In observing the specific cytokines induced in these studies, one might speculate a proinflammatory cascade phenomena. By way of example, TNF-a may result in the induction of IL-l#I, IL-6, and IL-8 [18]. TNF-a and/or IL-l are known to induce the synthesis and production of G-CSF [15]. Interleukin-10 serves to diminish the secretion of the aforementioned proinflammatory cytokines; namely TNF-a and IL-1B [ 181. Thus, the increased levels of IL-10 we observed may relate to a secondary regulatory mechanism. It is of interest to note that Thorn et al. reported increased serum levels of IL-6, IL-8 and intracellular adhesion molecule following hyperthermic limb perfusion with TNF-a, IFN-y, and melphalan [17]; this group found no detectable increase in IL-la, IL-@, IL-4, and leukemia inhibitory factor (LIF). Analysis of active TGF-/31 levels in radiant heat WBH patients suggests that no significant changes occur. A trend towards induction is observed in patients undergoing extra-corporeal WBH immediately after treatment at 41.8”C for 60 min. This was previously reported by Kekow et al. [3]. This apparent discrepancy between results obtained with radiant heat WBH versus extra-corporeal WBH may be technology dependent. Extra-corporeal WBH produces a destructive thrombocytopenia [12]. Platelets are a major source of TGF-/Il. Hence, the transient change of this cytokine seen during extra-corporeal WBH may simply relate to platelet disruption. Further investigation into a possible role for TGF-/I1 with WBH would probably be experimentally cumbersome; the synthesis, secretion, and activation of this cytokine is complex and tightly regulated. Hence, it is rapidly cleared from the blood. In this regard, measurement of the activated form of this molecule would require frequent sampling (e.g. 10 min intervals). Thus, further studies of this cytokine with WBH plasma samples are not currently planned. It is noteworthy that WBH has intrinsic cytoreductive potential. It is common in a number of ma-

Letters

97 (1995)

195-201

lignancies (e.g. leukemias, lymphomas, and malignant melanomas) to see dramatic but short-term remissions [ 121. Relevant to this, TNF-a [ 151 and IL-6 [ 181 are known to have significant anti-neoplastic activity. Based on these considerations, we are currently initiating pre-clinical investigations on the effects of’ hyperthermia and/or cytokines on cellular apoptosis and necrosis. In summary, the data reported here helps to explicate both laboratory and clinical results regarding hematological changes associated with WBH as an adjunct to cytotoxic therapy. Beyond this, these results may also have implications for other biological phenomena. It is conceivable that increased cytokine concentrations at a cellular level far exceed that which has been measured in the peripheral circulation. Our ongoing clinical investigations will include the assessment of other cytokines, as well as bone marrow samples derived from patients pre- and postWBH. Acknowledgements

We thank cussions and ported by the NY; Midwest Brookfield, MOlRR03186.

Dr. Lyold Old for his thoughtful dissuggestions. This research was supCancer Research Institute, New York, Athletes Against Childhood Cancer, WI; NIH UOlCA62491; GCRC-

References

VI d’Oleue, F., Robins, HI., Cohen, J.D., Schmitt, C.L. and

Spriggs, D. (1993) Cytokine induction in humans by 41XC whole body hypertbermia. J. Natl. Cancer Inst., 85, 833834. 121 Dinarello, C.A. and Wolff, SM. (1993) Mechanisms of disease: the role of interleukin-1 in disease. N. Engl. J. Med., 328, 106-113. t31 Kekow, J., Geisler, J., Szymkowiak, C., Wagner, T. and Wiedemann, G.J. (1995) Transforming growth factor beta induction by whole body hyperthermia (WBH). Am. Assoc. Cancer Res., 36,206 (Abstract 1227). [41 Lumpkin, M.D. (1987) The regulation of ACTH secretion by IL,.l. Science, 238,452-454. t51 Moore, M.A.S. and Warren, D.J. (1987) Synergy of interlet&in-l and granulocyte colony-stimulating factor: in vivo stimulation of stem-cell recovery and hematopoietic regeneration following 5-fluorouracil treatment of mice. Proc. Natl. Acad. Sci. USA, 84,7134-7138. 161 Robins, H.I., Schmitt-Tiggelaar, C.L., Cohen, J.D., Woods,

H.I. Robins et al. I Cancer

[7] [8]

[9]

[lo]

[11]

[12]

[13]

J.P., Heiss, C., Gillis, W. and dOleire, F. (1994) A new technological approach to radiant heat whole body hyperthennia. Cancer L&t., 79, 137-145. Robins, HI. (1995) Meeting Report: Systemic Hyperthermia Oncological Working Group. Oncology. Int. J. Cancer Res. Treat., in press. Robins, HI., Cohen, J.D., Schmitt, C.L., Tutsch, K.D., Feierabend, C., Arzoomanian, R.Z., Alberti, D., d’Oleite, F., Longo, W., Heiss, C., Rushing, D. and Spriggs, D. (1993) Phase I clinical trial of carboplatin and 41.8”C whole body hyperthermia in cancer patients. J. Clin. Oncol., 11, 17871794. Robins, HI., Kahn, N.H., Shelton, S.E., Martin, P.A., Shecterle, L.M., Barksdale, CM., Neville, A.J. and Marshall, J. (1987) Rise in plasma beta-endorphin, ACTH and cortisol in cancer patients undergoing whole body hyperthermia. Honn. Metab. Res., 19.441-443. Robins, H.I., Longo, W.L., Hugander, A., Bozdech, M., Schwartz, B., Steeves, R.A., Flynn, B., Triggs, M. and Sondel, P.M. (1986) Whole body hypertbermia combined with total body irradiation and chemotherapy as a preparative regimen for allogenic bone marrow transplantation in a patient with Burkitt’s lymphoma. Cancer J., 1, 180-183. Robins, HI., Longo, W.L., Steeves, R.A., Cohen, J.D., Schmitt, CL., Neville, A.J., O’Keefe, S., Lagoni, R. and Riggs, C. (1990) Adjunctive therapy (whole body hyperthermia versus lonidamine) to total body irradiation for the treatment of favorable B cell neoplasms: a report of two pilot clinical trials and laboratory investigations. Int. J. Radiat. Oncol. Biol. Phys., 18,909-920. Robins, H.I., Neville, A.J. and Cohen, J.D. (1992) Whole body hyperthermia: biological and clinical aspects. In: Clinical Thermology. Editor: M. Gautherie, SpringerVerlag, Berlin. Smith II, J.W., Longo, D.L., Alvord, W.G., Janik, J.E.,

Letters

[14] [15]

[16]

[17]

[18] [19]

[20]

97’ (1995)

195-201

201

Sharfman, W.H., Gause, B.L., Curti, B.D., Cmekmore, S.P., Holmlund, J.T., Fenton, R.G., Sznol, M., Miller, L.L., !Ihimizu, M., Oppenheim, J.J., Fiem, S.J., Hursey, J.C., Powers, G.C. and Urba, W.J. (1993) The effects of treatment with interleukin la on platelet recovery after high dose carboplatin. N. Engl. J. Med., 328,7X%761. Spangelo, B.L., Judd, A.M. and Is&son, P.C. (1989) IL-6 stimulates anterior pituitary hormone release in vitro. Endocrinology, 125,575-577. !ipriggs, D.R. and Yates, S.W. (1992) Cancer chemotherapy, experiences with TNF administration in humans. In: Tumor Necrosis Factors: the molecules and their emerging role in medicine, pp. 383406. Editor: B. Beutler. Raven Press, New York. Steeves,R., Robins, H.I., Miller, K., Shecterle, L., Kruk, M. :md Dennis, W. (1987) Interaction of whole body hyperthermia and irradiation in the treatment of AKR mouse leukemia. Int. J. Radiat. Biol., 52,935-947. Thorn, A.K., Alexander, H.R., Andrich, M.P., Barker, W.C., ‘Rosenberg, S.A. and Fraker, V.L. (1995) Cytokine levels