- Email: [email protected]
The effect of chloroquine and hyperthermia on murine neuroblastoma
The Effect of Chloroquine By Roland Thomas,
and Hyperthermia Dennis W. Vane,
Jay L. Grosfeld,
Indianapolis, 0 A number of reports suggest that hyperthermia is an adjunctive treatment modality in management of neural crest tumors. Recent studies have demonstrated a synergistic effect of induced hyperthermia when coupled with chloroquine in an in vitro model. This study examines the effect of chloroquine and hyperthermia in an in vivo murine neuroblastoma model. Forty-seven Ajax white mice (weighing 20 to 30 g) received a subaxillary tumor burden (C-1300 murine neuroblastoma) per trochar (1.25 x IO’ cells). The tumor was then incubated for 9 days. Mice were then divided into four groups: group 1, controls (n = 15): group 2, hyperthermia In = 12); group 3. chloroquine (n = 10); and group 4, chloroquine with hyperthermia (n = IO). Hyperthermia was induced with 40 to 59 mW/ cm’ at 2,450 MHz microwave radiation for 4 minutes to achieve a temperature of 41.5”C for IO of 14 treatment days. Chloroquine was administered intraperitoneally at a dose of 40 mg/kg body weight for IO of 14 treatment days. Mice were weighed and tumor size was determined daily. Animals were killed on day 21 and postmortem examination was performed, with tumors graded histologically. Animal weight, tumor weight, and tumor size were similar for all groups (P > .05). Mortality was 6% in group I, 25% in group 2, 50% in group 3, and 40% in group 4 (P < .05). Rate of tumor metastases was also statistically different from controls: group 1, 0%: group 2. 60%; group 3, 90%: and group 4, 90% (P < .05). Chloroquine acts by altering lysosomal function and as a DNA/RNA polymerase inhibitor. In vitro this mechanism apparently allows for increased tumor kill when hyperthermia is added. However, in the in vivo model, inhibition of host rapid cell replication (reticuloendothelial system, and white cell production) by chloroquine may explain the increased incidence of metastatic spread and mortality. In addition, microwave-induced hyperthermia alone (previously reported to reduce tumor burden) when administered in a repeated fashion also resulted in an increased metastatic rate. This may be due to increased tumor perfusion and seeding as a result of a chronically induced hyperemic state. @ 1990 by W.B. Saunders Company.
effective
INDEX WORD: Neuroblastoma.
D
ESPITE DRAMATIC advances in cancer therapy, overall prognosis for children with neuroblastoma has remained unchanged for the past 20 years. This study seeks to investigate the effects of hyperthermia and chloroquine on murine neuroblastoma. Hyperthermia has been a useful adjunct in the management of several tumors, including neuroblastoma. Chloroquine, commonly used as an antimalarial agent, has recently received renewed attention in cancer therapy because of its ability to inhibit DNA polymerase. Recent reports have shown efficacy of chloroquine and hyperthermia on melanoma in vitro.’ Neuroblastoma is an embryonal APUD tumor of neural-crest origin Journsf of Pediatric Surgery,
Vol 25, No 9 (September),
1990: pp 929-932
on Murine Neuroblastoma and Philip R. Faught
Indiana
like melanoma, with known sensitivity to hyperthermic treatment. The purpose of this study is to evaluate the effect of chloroquine and hyperthermia on murine neuroblastoma in vivo. MATERIALS
AND METHODS
C-1300 murine neuroblastoma was transplanted into 52 male white Ajax mice (weight, 20 to 30 g). Tumor plugs (1.25 x lo6 cells) were placed subcutaneously in an axillary position with standard trochar technique and the tumor was allowed to incubate for 9 days. Mice were then divided randomly into four groups: group 1, controls (n = 20); group 2, treatment with hyperthermia only (n = 12); group 3, treatment with chloroquine only (n = 10); and group 4, treatment with chloroquine and hyperthermia (n = LO). Mice treated with chloroquine received 40 mg/kg chloroquine intraperitoneally for 10 of 14 days. Mice treated with hyperthermia were given 40 to 60 mW/cm* at 2,450 MHz microwave radiation for 4 minutes (Elmed diathermy unit; Elmed Inc, Addison, IL) to achieve a core temperature of 41.5OC again for 10 of 14 days. Animals receiving both modalities were injected with chloroquine 30 minutes prior to microwave treatment. Temperatures were measured with a BAT-8 thermistor probe (Bailey Instruments Inc, Saddle Brook, NJ) inserted intramuscularly in the mouse’s hind quarter immediately after microwave treatment. All mice were weighed and tumor size measured daily. Tumor volumes were measured in standard fashion: V = (length x width x height).2 All tumors were examined histologically for tumor necrosis. All assays were performed by an independent pathologist with blinded specimens. Tumors were graded from 1 to 6. One stood for minimal or no necrosis of the specimen, 2 for less than 25% necrosis, 3 for 25% to 50% necrosis, 4 for 50% to 75% necrosis, 5 for more than 75% necrosis, and 6 for total or near total necrosis. RESULTS
Animal
Weight
At treatment day 0, (day 9 after transplantation) average animal weight was: group 1, 23.20 + .05 g; group 2, 23.25 + .05 g; group 3, 22.95 + .05 g; and group 4, 22.05 + .05 g. Weights decreased at differing rates as shown in Fig 1 and then increased to day 14. Final weights were: group 1, 21.60 r .05 g; group 2, 21.60 + .05 g; group 3, 22.50 + .05 g; and group 4, From the Section of Pediatric Surgery, Department of Surgery, and the Section of Pediatric Pathology, Indiana University Medical Cenrer and the James Whitcomb Riley Hospital for Children, Indianapolis, IN. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics. Chicago, Illinois, October 21-23, 1989. Address reprint requests to Jay L. Grosfeld, MD, Surgeon-inChief. J. W. Riley Hospitalfor Children, 702 Barnhill Dr. Indianapolis, IN 46202-5200. 8 1990 by W.B. Saunders Company. 0022-3468/90/2509-0002$03.00/0
930
THOMAS
23
ET AL
7
Q W t
21
9
r a m
19
CONTROL
s
CHLORO
C&H
Fig 2. Tumor volumes on day 0 (WI for each group ware statistically identical. On day 14 @I all groups ware-statistically larger than they were at day 0. Groups 2 and 4 were smaller than group 1 at dey 14 and group 3 was stetirtically identical to group 1. lP 4 .Ol vcontrol.
17
15
HYPER
I
1 2
I
I
/
I,
3
4
5
6
I
7
8
I
,
I
,
/
,
1
9 10 11 12131415
Treat men t day Fig 1. Animal weight at nadir appaared about day 7 and ware all statistically smaller than day 0. Groups 1 and 2 remained smaller than their day 0 weights at the conclusion of the expariment. ., control; +. hyperthermia: l, chloroquine: 0, chloroquine and hyperthermia.
21.50 f .05 g. Weight differences at nadir were statistically different for all four groups from day 0 (P c .05); however, only groups 1 and 2 achieved significance (P < .OS) on day 14 when compared with day 0. Tumor Volume The tumor in untreated animals was palpable 5 days after transplantation. Tumor volume was calculated as described previously. Depth of tumor was easily measured because the tumor was freely mobile in its subcutaneous location. At treatment day 0 tumor volumes were: group 1,451 * 30 mm3; group 2,437 * 30 mm3; group 3,520 + 30 mm3; and group 4,53 1 f 30 mm3. At the conclusion of treatment (day 14) volumes were: group 1, 8,158 f 30 mm3; group 2, 5,479 f 30 mm3; group 3, 8,250 f 30 mm3; and group 4, 7,145 * 30 mm3 (Fig 2). Initial tumor volumes (day 0) were not statistically different from each other; however, on day 14 all tumors .were significantly larger than their original sizes (P-z .OOl). At day 14, group 2 was significantly smaller than group 1 (control) (P -C .Ol) as was group 4 (P < .Ol). Group 3 was statistically identical to the controls (group 1) (Fig 2). Tumor Weight Tumor weight was initially measured as. 157 + .Ol g at day 0. Tumor weight at day 14 of treatment was: group 1, 4.6 + .10 g; group 2, 3.9 + .10 g; group 3,
6.3 + .I0 g; and group 4, 5.5 * .lO g. All groups were statistically larger at day 14 than day 0 (P < .OOl). On day 14, group 2 was statistically smaller than controls (P < .05); both groups 3 and 4 were statistically larger than the control group (P -C.Ol) (Fig 3). Mortality Mortality was determined by ascertaining the number of experimental animals that did not reach day 14 of treatment. Mortality for each group was: group 1, 6% (n = 1); group 2, 25% (n = 3); group 3, 50% (n = 5); and group 4, 40% (n = 4). All treatment groups had a significantly higher mortality than the control group (P-C .05). Mortality in group 3 was higher than group 2 (P < .05) but group 4 was not statistically higher than group 2 (P > .05) (Fig 4). Histoiogical Anaiysis All tumors were graded in a blinded fashion by an independent pathologist. On day 0 tumors were read as having an average histological score of 1.7. Group 1 had a score of 4.4; group 2,4.6; group 3,4.9; and group
I
II
IV
Ill
GROUP =
CONTROL
I
CHLOR
lo 111
@&% CONTROL mCdH
I
0
HYPER
II
I”
Fig 3. Tumor weight were statistically heavier than day 0 on day 14. Group 2 was statistically smaller than controlr (group 1) while both groups 3 end 4 ware statistically heavier. lP c .06 v controls. l lf -z .06 v group 1.
931
CHLOROQUINE AND HYPERTHERMIA IN NEUROBLASTOMA 60 50
i
r*-40 %
30 20
Ill
II
IV
GROUP Fig 4. All treatment groups had significantly higher mortality than the control group. Mortality in group 3 was higher than group 2 but group 4 was not statistically increased over group 2. W. Control: @, hyperthermia: q, chloroquine; k% chloroquine and hyperthermia. .P rc .06 Y group 2.
4, 4.6. Although statistics cannot be generated from these types of data it was the opinion of the pathologist that group 3 exhibited significantly more necrosis than the other groups and all treated groups showed more necrosis than the controls. Metastases
All animals were killed at the conclusion of 14 days of treatment (day 15). All underwent complete autopsy and were examined for invasion of tumor and metastatic spread. No animals in group 1 had any evidence of metastatic spread or tumor invasion. Interestingly, six animals in group 2 had metastases to the liver. Nine animals in both groups 3 and 4 had metastatic spread, and in both cases metastases were found in both liver and lung. In all animals with metastatic spread, metastases were multiple in the affected organ. DISCUSSION
Survival in children with neuroblastoma has not improved significantly during the past 25 years.3 Previous studies have determined a beneficial role of hyperthermia in the response of this tumor, although presently this modality is not part of the clinical regimen.4 Recent publications have identified the drug chloroquine (an antimalarial agent) as being a potent enhancement agent in vitro for hyperthermia when used to treat melanoma, also known to be hyperthermiasensitive. ’ Neuroblastoma and melanoma are APUD tumors of neural crest origin with some similar characteristics like their sensitivity to hyperthermia.’ Chloroquine is a potent inhibitor of DNA polymerase and strongly inhibits both DNA and RNA synthesis.6 Its action is directly against the repair of single-strand DNA breaks caused by a variety of agents. In addition, chloroquine directly inhibits both chemotaxis and phagocytosis of
the polymorphonuclear lymphocyte, dramatically decreasing inflamatory response.7 Unfortunately this action is very poorly understood and no clear explanation of this phenomena is available. In spite of the remarkable response of melanoma to hyperthermia and chloroquine treatment, similar response by neuroblastoma could not be elicited.’ In this study, neuroblastoma grew faster, spread more rapidly, and caused a higher mortality when chloroquine was added to hyperthermia as a treatment agent. Several hypotheses may explain these differences. Chloroquine, although its actions are poorly understood, has a propensity to adhere to human melanin cells.7 This adherence does not seem to effect normal cells except that hyperpigmentation is stimulated. It is possible that adherence makes the cell more sensitive to external factors such as increased heat sensitivity, but no data are available in this area. Hyperthermia has numerous effects on tumor physiology.’ It is reported to stimulate oxygen disturbances in the tumor tissue by causing functional disorders in the blood flow. In this same fashion, heat dissipation from the neoplasm is impaired, which further increases the kill rate on the tissue? In addition to all these factors, hyperthermia causes functional disturbances of the cell metabolism, which include breakdown of amino acid structures and tissue acidosis.” It is this mechanism that chloroquine should theoretically enhance to stimulate greater tumor cell destruction. Chloroquine attacks single strand DNAase, which should impede the repair of damaged DNA caused by hyperthermia induction.” Murine C- 1300 neuroblastoma is sensitive to hyperthermia, particularly when coupled with a phosphodiesterase inhibitor such as papavarine.4 If this action is through inability of the malignant cell to repair itself after damage from induced thermal radiation, then subsequently, drugs such as chloroquine (a potent inhibitor of repair) should react similarly. I2 However, this was not the case in the present study because chloroquine had an adverse effect on tumor treatment. In addition, other studies combining chloroquine with VP16-312 (known to produce single-strand DNA breaks) similarly failed to demonstrate enhancement of tumor cell kill in an in vivo model.6 It is apparent that some other effect of chloroquine may be a factor that prevents tumor destruction in this setting. The most likely explanation is the inhibition of phagocytosis and inflammation caused by the drug.’ Stabilization of the lysosomal membrane of the polymorphonuclear cells by chloroquine may render damaged tumor cells immune to the body’s normal protective systems. This would explain why chloroquine is effective in vitro and unsuccessful in vivo.‘q6
932
THOMAS ET AL
This report clearly supports the beneficial effects of hyperthermia in reducing tumor growth and response. Chloroquine was not effective in increasing tumor response and resulted in an increased animal mortality
associated with tumor growth and metastasis. Actions of chloroquine, other than its ability to inhibit DNAase, appear to be responsible for its lack of tumoricidal effect in an in vitro model.
REFERENCES 1. Morrow M, Hager C, Berger D, et al: Chloroquine as a hyperthermia potentiator. J Surg Res 46637-639, 1989 2. Okuzono S, Nakagawara A, Sue K, et al: Different clonal drug sensitivity of murine neuroblastoma cells in-vivo. J Pediatr Surg 23:962-699,198s 3. Vane DW, Grosfeld JL: Solid malignant tumors in children: An experience with 506 cases. Bull Sot Sci Med Luxembourg 197-216, 1987 4. West KW, Weber TR, Grosfeld JL: Synergistic effect of hyperthermia, papaverine, and chemotherapy in murine neuroblastoma. J Pediatr Surg 15:913-917, 1980 5. Grosfeld JL: Neuroblastoma: Current concepts of management, Surg Rounds 10:47-60.1987 6. Arnold AM, Whitehouse JMA: Interaction of VP16-213 with the DNA repair antagonist chloroquine. Cancer Chemother Pharmaco1 7:123-126, 1982
7. Mackenzie A: An appraisal of chloroquine. Arth Rheumatol 13:280-291,197O 8. Otte J: Hyperthermia 147:560-569,1988
in cancer therapy.
Eur J Pediatr
9. Vaupel P, Kallinowski F, Kluge M: Pathophysiology of tumors in hyperthermia. Ret Results Cancer Res 107:65-75.1988 10. Oleson JR, Calderwocd SK, Coughlin CT, et al: Biological and clinical aspects of hyperthermia in cancer therapy. Am J Clin Oncol 11:368-380, 1988 11. Streff C: Aspects of metabolic change after hyperthermia. Ret Results Cancer Res 107:7-15, 1988 12. Rama BN, Prasad KN: Modification of the hyperthermic response on neuroblastoma cells by CAMP and sodium butyrate. Cancer 58:1448-1452,1986
Discussion S. Shochat
(Stanford,
CA): I take it that this was
local hyperthermia rather than systemic and in a tumor like neuroblastoma, I wonder whether that’s going to be practical. I also wonder whether your results may have turned out the way they did because of the particular C-l 300 tumor that you used. Did you try other clones or strains of neuroblastoma? Did you consider other tumors other than neuroblastoma? R. Thomas (response): Actually, the animals re-
ceived whole-body microwave irradiation and totalbody warming, and the temperatures measured were central temperatures. We have not yet examined the effect of either hyperthermia or chloroquine and hyperthermia on strains of neuroblastoma other than C1300. Given the known effect on in vitro melanoma, it would be interesting to see what the effect of chloroquine and hyperthermia is on an in vivo melanoma model.
and Hyperthermia Dennis W. Vane,
Jay L. Grosfeld,
Indianapolis, 0 A number of reports suggest that hyperthermia is an adjunctive treatment modality in management of neural crest tumors. Recent studies have demonstrated a synergistic effect of induced hyperthermia when coupled with chloroquine in an in vitro model. This study examines the effect of chloroquine and hyperthermia in an in vivo murine neuroblastoma model. Forty-seven Ajax white mice (weighing 20 to 30 g) received a subaxillary tumor burden (C-1300 murine neuroblastoma) per trochar (1.25 x IO’ cells). The tumor was then incubated for 9 days. Mice were then divided into four groups: group 1, controls (n = 15): group 2, hyperthermia In = 12); group 3. chloroquine (n = 10); and group 4, chloroquine with hyperthermia (n = IO). Hyperthermia was induced with 40 to 59 mW/ cm’ at 2,450 MHz microwave radiation for 4 minutes to achieve a temperature of 41.5”C for IO of 14 treatment days. Chloroquine was administered intraperitoneally at a dose of 40 mg/kg body weight for IO of 14 treatment days. Mice were weighed and tumor size was determined daily. Animals were killed on day 21 and postmortem examination was performed, with tumors graded histologically. Animal weight, tumor weight, and tumor size were similar for all groups (P > .05). Mortality was 6% in group I, 25% in group 2, 50% in group 3, and 40% in group 4 (P < .05). Rate of tumor metastases was also statistically different from controls: group 1, 0%: group 2. 60%; group 3, 90%: and group 4, 90% (P < .05). Chloroquine acts by altering lysosomal function and as a DNA/RNA polymerase inhibitor. In vitro this mechanism apparently allows for increased tumor kill when hyperthermia is added. However, in the in vivo model, inhibition of host rapid cell replication (reticuloendothelial system, and white cell production) by chloroquine may explain the increased incidence of metastatic spread and mortality. In addition, microwave-induced hyperthermia alone (previously reported to reduce tumor burden) when administered in a repeated fashion also resulted in an increased metastatic rate. This may be due to increased tumor perfusion and seeding as a result of a chronically induced hyperemic state. @ 1990 by W.B. Saunders Company.
effective
INDEX WORD: Neuroblastoma.
D
ESPITE DRAMATIC advances in cancer therapy, overall prognosis for children with neuroblastoma has remained unchanged for the past 20 years. This study seeks to investigate the effects of hyperthermia and chloroquine on murine neuroblastoma. Hyperthermia has been a useful adjunct in the management of several tumors, including neuroblastoma. Chloroquine, commonly used as an antimalarial agent, has recently received renewed attention in cancer therapy because of its ability to inhibit DNA polymerase. Recent reports have shown efficacy of chloroquine and hyperthermia on melanoma in vitro.’ Neuroblastoma is an embryonal APUD tumor of neural-crest origin Journsf of Pediatric Surgery,
Vol 25, No 9 (September),
1990: pp 929-932
on Murine Neuroblastoma and Philip R. Faught
Indiana
like melanoma, with known sensitivity to hyperthermic treatment. The purpose of this study is to evaluate the effect of chloroquine and hyperthermia on murine neuroblastoma in vivo. MATERIALS
AND METHODS
C-1300 murine neuroblastoma was transplanted into 52 male white Ajax mice (weight, 20 to 30 g). Tumor plugs (1.25 x lo6 cells) were placed subcutaneously in an axillary position with standard trochar technique and the tumor was allowed to incubate for 9 days. Mice were then divided randomly into four groups: group 1, controls (n = 20); group 2, treatment with hyperthermia only (n = 12); group 3, treatment with chloroquine only (n = 10); and group 4, treatment with chloroquine and hyperthermia (n = LO). Mice treated with chloroquine received 40 mg/kg chloroquine intraperitoneally for 10 of 14 days. Mice treated with hyperthermia were given 40 to 60 mW/cm* at 2,450 MHz microwave radiation for 4 minutes (Elmed diathermy unit; Elmed Inc, Addison, IL) to achieve a core temperature of 41.5OC again for 10 of 14 days. Animals receiving both modalities were injected with chloroquine 30 minutes prior to microwave treatment. Temperatures were measured with a BAT-8 thermistor probe (Bailey Instruments Inc, Saddle Brook, NJ) inserted intramuscularly in the mouse’s hind quarter immediately after microwave treatment. All mice were weighed and tumor size measured daily. Tumor volumes were measured in standard fashion: V = (length x width x height).2 All tumors were examined histologically for tumor necrosis. All assays were performed by an independent pathologist with blinded specimens. Tumors were graded from 1 to 6. One stood for minimal or no necrosis of the specimen, 2 for less than 25% necrosis, 3 for 25% to 50% necrosis, 4 for 50% to 75% necrosis, 5 for more than 75% necrosis, and 6 for total or near total necrosis. RESULTS
Animal
Weight
At treatment day 0, (day 9 after transplantation) average animal weight was: group 1, 23.20 + .05 g; group 2, 23.25 + .05 g; group 3, 22.95 + .05 g; and group 4, 22.05 + .05 g. Weights decreased at differing rates as shown in Fig 1 and then increased to day 14. Final weights were: group 1, 21.60 r .05 g; group 2, 21.60 + .05 g; group 3, 22.50 + .05 g; and group 4, From the Section of Pediatric Surgery, Department of Surgery, and the Section of Pediatric Pathology, Indiana University Medical Cenrer and the James Whitcomb Riley Hospital for Children, Indianapolis, IN. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics. Chicago, Illinois, October 21-23, 1989. Address reprint requests to Jay L. Grosfeld, MD, Surgeon-inChief. J. W. Riley Hospitalfor Children, 702 Barnhill Dr. Indianapolis, IN 46202-5200. 8 1990 by W.B. Saunders Company. 0022-3468/90/2509-0002$03.00/0
930
THOMAS
23
ET AL
7
Q W t
21
9
r a m
19
CONTROL
s
CHLORO
C&H
Fig 2. Tumor volumes on day 0 (WI for each group ware statistically identical. On day 14 @I all groups ware-statistically larger than they were at day 0. Groups 2 and 4 were smaller than group 1 at dey 14 and group 3 was stetirtically identical to group 1. lP 4 .Ol vcontrol.
17
15
HYPER
I
1 2
I
I
/
I,
3
4
5
6
I
7
8
I
,
I
,
/
,
1
9 10 11 12131415
Treat men t day Fig 1. Animal weight at nadir appaared about day 7 and ware all statistically smaller than day 0. Groups 1 and 2 remained smaller than their day 0 weights at the conclusion of the expariment. ., control; +. hyperthermia: l, chloroquine: 0, chloroquine and hyperthermia.
21.50 f .05 g. Weight differences at nadir were statistically different for all four groups from day 0 (P c .05); however, only groups 1 and 2 achieved significance (P < .OS) on day 14 when compared with day 0. Tumor Volume The tumor in untreated animals was palpable 5 days after transplantation. Tumor volume was calculated as described previously. Depth of tumor was easily measured because the tumor was freely mobile in its subcutaneous location. At treatment day 0 tumor volumes were: group 1,451 * 30 mm3; group 2,437 * 30 mm3; group 3,520 + 30 mm3; and group 4,53 1 f 30 mm3. At the conclusion of treatment (day 14) volumes were: group 1, 8,158 f 30 mm3; group 2, 5,479 f 30 mm3; group 3, 8,250 f 30 mm3; and group 4, 7,145 * 30 mm3 (Fig 2). Initial tumor volumes (day 0) were not statistically different from each other; however, on day 14 all tumors .were significantly larger than their original sizes (P-z .OOl). At day 14, group 2 was significantly smaller than group 1 (control) (P -C .Ol) as was group 4 (P < .Ol). Group 3 was statistically identical to the controls (group 1) (Fig 2). Tumor Weight Tumor weight was initially measured as. 157 + .Ol g at day 0. Tumor weight at day 14 of treatment was: group 1, 4.6 + .10 g; group 2, 3.9 + .10 g; group 3,
6.3 + .I0 g; and group 4, 5.5 * .lO g. All groups were statistically larger at day 14 than day 0 (P < .OOl). On day 14, group 2 was statistically smaller than controls (P < .05); both groups 3 and 4 were statistically larger than the control group (P -C.Ol) (Fig 3). Mortality Mortality was determined by ascertaining the number of experimental animals that did not reach day 14 of treatment. Mortality for each group was: group 1, 6% (n = 1); group 2, 25% (n = 3); group 3, 50% (n = 5); and group 4, 40% (n = 4). All treatment groups had a significantly higher mortality than the control group (P-C .05). Mortality in group 3 was higher than group 2 (P < .05) but group 4 was not statistically higher than group 2 (P > .05) (Fig 4). Histoiogical Anaiysis All tumors were graded in a blinded fashion by an independent pathologist. On day 0 tumors were read as having an average histological score of 1.7. Group 1 had a score of 4.4; group 2,4.6; group 3,4.9; and group
I
II
IV
Ill
GROUP =
CONTROL
I
CHLOR
lo 111
@&% CONTROL mCdH
I
0
HYPER
II
I”
Fig 3. Tumor weight were statistically heavier than day 0 on day 14. Group 2 was statistically smaller than controlr (group 1) while both groups 3 end 4 ware statistically heavier. lP c .06 v controls. l lf -z .06 v group 1.
931
CHLOROQUINE AND HYPERTHERMIA IN NEUROBLASTOMA 60 50
i
r*-40 %
30 20
Ill
II
IV
GROUP Fig 4. All treatment groups had significantly higher mortality than the control group. Mortality in group 3 was higher than group 2 but group 4 was not statistically increased over group 2. W. Control: @, hyperthermia: q, chloroquine; k% chloroquine and hyperthermia. .P rc .06 Y group 2.
4, 4.6. Although statistics cannot be generated from these types of data it was the opinion of the pathologist that group 3 exhibited significantly more necrosis than the other groups and all treated groups showed more necrosis than the controls. Metastases
All animals were killed at the conclusion of 14 days of treatment (day 15). All underwent complete autopsy and were examined for invasion of tumor and metastatic spread. No animals in group 1 had any evidence of metastatic spread or tumor invasion. Interestingly, six animals in group 2 had metastases to the liver. Nine animals in both groups 3 and 4 had metastatic spread, and in both cases metastases were found in both liver and lung. In all animals with metastatic spread, metastases were multiple in the affected organ. DISCUSSION
Survival in children with neuroblastoma has not improved significantly during the past 25 years.3 Previous studies have determined a beneficial role of hyperthermia in the response of this tumor, although presently this modality is not part of the clinical regimen.4 Recent publications have identified the drug chloroquine (an antimalarial agent) as being a potent enhancement agent in vitro for hyperthermia when used to treat melanoma, also known to be hyperthermiasensitive. ’ Neuroblastoma and melanoma are APUD tumors of neural crest origin with some similar characteristics like their sensitivity to hyperthermia.’ Chloroquine is a potent inhibitor of DNA polymerase and strongly inhibits both DNA and RNA synthesis.6 Its action is directly against the repair of single-strand DNA breaks caused by a variety of agents. In addition, chloroquine directly inhibits both chemotaxis and phagocytosis of
the polymorphonuclear lymphocyte, dramatically decreasing inflamatory response.7 Unfortunately this action is very poorly understood and no clear explanation of this phenomena is available. In spite of the remarkable response of melanoma to hyperthermia and chloroquine treatment, similar response by neuroblastoma could not be elicited.’ In this study, neuroblastoma grew faster, spread more rapidly, and caused a higher mortality when chloroquine was added to hyperthermia as a treatment agent. Several hypotheses may explain these differences. Chloroquine, although its actions are poorly understood, has a propensity to adhere to human melanin cells.7 This adherence does not seem to effect normal cells except that hyperpigmentation is stimulated. It is possible that adherence makes the cell more sensitive to external factors such as increased heat sensitivity, but no data are available in this area. Hyperthermia has numerous effects on tumor physiology.’ It is reported to stimulate oxygen disturbances in the tumor tissue by causing functional disorders in the blood flow. In this same fashion, heat dissipation from the neoplasm is impaired, which further increases the kill rate on the tissue? In addition to all these factors, hyperthermia causes functional disturbances of the cell metabolism, which include breakdown of amino acid structures and tissue acidosis.” It is this mechanism that chloroquine should theoretically enhance to stimulate greater tumor cell destruction. Chloroquine attacks single strand DNAase, which should impede the repair of damaged DNA caused by hyperthermia induction.” Murine C- 1300 neuroblastoma is sensitive to hyperthermia, particularly when coupled with a phosphodiesterase inhibitor such as papavarine.4 If this action is through inability of the malignant cell to repair itself after damage from induced thermal radiation, then subsequently, drugs such as chloroquine (a potent inhibitor of repair) should react similarly. I2 However, this was not the case in the present study because chloroquine had an adverse effect on tumor treatment. In addition, other studies combining chloroquine with VP16-312 (known to produce single-strand DNA breaks) similarly failed to demonstrate enhancement of tumor cell kill in an in vivo model.6 It is apparent that some other effect of chloroquine may be a factor that prevents tumor destruction in this setting. The most likely explanation is the inhibition of phagocytosis and inflammation caused by the drug.’ Stabilization of the lysosomal membrane of the polymorphonuclear cells by chloroquine may render damaged tumor cells immune to the body’s normal protective systems. This would explain why chloroquine is effective in vitro and unsuccessful in vivo.‘q6
932
THOMAS ET AL
This report clearly supports the beneficial effects of hyperthermia in reducing tumor growth and response. Chloroquine was not effective in increasing tumor response and resulted in an increased animal mortality
associated with tumor growth and metastasis. Actions of chloroquine, other than its ability to inhibit DNAase, appear to be responsible for its lack of tumoricidal effect in an in vitro model.
REFERENCES 1. Morrow M, Hager C, Berger D, et al: Chloroquine as a hyperthermia potentiator. J Surg Res 46637-639, 1989 2. Okuzono S, Nakagawara A, Sue K, et al: Different clonal drug sensitivity of murine neuroblastoma cells in-vivo. J Pediatr Surg 23:962-699,198s 3. Vane DW, Grosfeld JL: Solid malignant tumors in children: An experience with 506 cases. Bull Sot Sci Med Luxembourg 197-216, 1987 4. West KW, Weber TR, Grosfeld JL: Synergistic effect of hyperthermia, papaverine, and chemotherapy in murine neuroblastoma. J Pediatr Surg 15:913-917, 1980 5. Grosfeld JL: Neuroblastoma: Current concepts of management, Surg Rounds 10:47-60.1987 6. Arnold AM, Whitehouse JMA: Interaction of VP16-213 with the DNA repair antagonist chloroquine. Cancer Chemother Pharmaco1 7:123-126, 1982
7. Mackenzie A: An appraisal of chloroquine. Arth Rheumatol 13:280-291,197O 8. Otte J: Hyperthermia 147:560-569,1988
in cancer therapy.
Eur J Pediatr
9. Vaupel P, Kallinowski F, Kluge M: Pathophysiology of tumors in hyperthermia. Ret Results Cancer Res 107:65-75.1988 10. Oleson JR, Calderwocd SK, Coughlin CT, et al: Biological and clinical aspects of hyperthermia in cancer therapy. Am J Clin Oncol 11:368-380, 1988 11. Streff C: Aspects of metabolic change after hyperthermia. Ret Results Cancer Res 107:7-15, 1988 12. Rama BN, Prasad KN: Modification of the hyperthermic response on neuroblastoma cells by CAMP and sodium butyrate. Cancer 58:1448-1452,1986
Discussion S. Shochat
(Stanford,
CA): I take it that this was
local hyperthermia rather than systemic and in a tumor like neuroblastoma, I wonder whether that’s going to be practical. I also wonder whether your results may have turned out the way they did because of the particular C-l 300 tumor that you used. Did you try other clones or strains of neuroblastoma? Did you consider other tumors other than neuroblastoma? R. Thomas (response): Actually, the animals re-
ceived whole-body microwave irradiation and totalbody warming, and the temperatures measured were central temperatures. We have not yet examined the effect of either hyperthermia or chloroquine and hyperthermia on strains of neuroblastoma other than C1300. Given the known effect on in vitro melanoma, it would be interesting to see what the effect of chloroquine and hyperthermia is on an in vivo melanoma model.