- Email: [email protected]
Chemosensitivity testing of nonsolid tumors by the subrenal-capsule implant assay
GYNECOLOGIC
ONCOLOGY
17, 185-188 (1984)
Chemosensitivity Testing of Nonsolid Tumors by the Subrenal-Capsule Implant Assay JOAN
A. STRATTON, Ph.D.,’ JOHN P. MICHA, M.D., MARK A. RETTENMAIER, M.D., PATRICIA S. BRALEY, M.D., AND PHILIP J. DISAIA, M.D.
University
of California,
Irvine Medical Center, Department of Obstetrics and Gynecology, IO1 City Drive, Orange, California 92668
Received August 18, 1982 A procedure to test chemosensitivity of nonsolid tumors using the subrenal-capsule xenograft assay has been developed. This technique is applicable to the study of malignant effusions and hematopoietic tumors. The tumor cells are centrifuged to form a pellet and the pellet is resuspended in the presence of bovine fibrinogen to form a jelled clot. This clot is cut into fragments which are implanted beneath the renal capsule of normal mice. The growth of the tumor cells from ovarian adenocarcinoma obtained from ascitic fluid is better than that of tumor fragments from the solid tumors. This is probably a reflection of the greater viability of tumor cells from ascitic fluids. The sensitivity to chemotherapy was the same for the solid tumor and its malignant effusions.
INTRODUCTION The subrenal-capsule assay method of determining chemosensitivity of human tumors was developed by Arthur Bogden and colleagues [l] at the Mason Research Institute in Worcester, Mass. This technique has the advantages of being an in vivo assay and any metabolic steps required for activation of chemotherapeutic drugs are performed by the intact animal. The assay provides employable information in a short period of time so that initiation of chemotherapy is not delayed. Since each tumor fragment serves as its own control, both oncolysis and oncostasis can be determined. And finally, combination chemotherapies can be assayed. We have found an 84% long-term positive correlation with the results of the assay and the response of patients with various gynecologic malignancies to chemotherapy [3]. This manuscript deals with solid tumors and the long-term (12 months) clinical follow-ups and validates the subrenal-capsule tumor implant assay as a preclinical predictor of clinical sensitivity to chemotherapeutic agents. To date only solid tumor explants have been studied in the subrenal-capsule xenograft assay. In this study, we report an adaptation of this assay which allows ’ To whom requests for reprints and correspondence should be addressed. 185 OWO-8258/84 $1.50 Copyright 0 1984 by Academic Press, Inc. All rights of reproduction in any form reserved.
186
STRATTON
ET AL.
chemosensitivity testing of malignant cell suspensions. Thus the chemosensitivity of the patient’s tumor may be followed without further surgery. MATERIALS AND METHODS Culture medium. Minimum Eagle’s medium (MEM, Grand Island Biological Co., Oakland, Calif.) containing 10,000 units each of penicillin and streptomycin (penicillin-streptomycin, GIBCO), 0.05 mg/ml gentamycin sulfate (Garamycin, Shering Corp., Kenilwood, N.J.), 2 mM glutamine (GIBCO), and 20% newborn calf serum (NBS, GIBCO). Single cell suspensions. The human myeloid leukemia cell line K562, obtained from S. H. Golub (Department of Surgery, University of California at Los Angeles) has been maintained in culture for several years. The viability (trypan-blue exclusion) and the number of tumor cells were determined by counting in a hemocytometer. Cell suspensions were prepared containing 1 x 107, 2 x 106, and 2 x lo5 live K.562 cells. The cell suspensions were centrifuged at 1200 rpm for 10 mitt, the culture medium was discarded, and the cell pellet was resuspended in 1 ml of MEM and 20% NBS. Ten milligrams of bovine fibrinogen (F-4000, Sigma Chemical Co., St. Louis, MO.) was added to the resuspended cells and the suspension centrifuged at 1000 rpm for 10 min. The pellet was incubated at 37°C for lo-30 min until a firm clot formed. The jelled cell pellet was poured into a cold petri dish with fresh cold MEM and 10% dextrose ((Y- D( +) glucose, Sigma), and then cut into small pieces (approximately 1 mm3). The fragments tend to be sticky; the dextrose helps to keep them separated. The fragments were implanted beneath the renal capsule of normal BDF, mice as described previously for solid tumors [ 1,2]. Malignant ascites was obtained by paracentesis from four patients with ovarian adenocarcinoma. Solid tumor was also obtained from two of these patients at laparotomy. The ascites was allowed to clot naturally, and the clot and sedimented cells were centrifuged at 1200 t-pm for 15 min. Firmer clots were obtained if the ascitic fluid was allowed to settle overnight at 4°C. If the natural clots were too friable, 10 mg of bovine fibrinogen was added to the vortexed clot, and the suspension was centrifuged and handled as described for the K562 leukemia. If no natural clot formed, the ascitic fluid was centrifuged at 1200 rpm for 15 mitt, the cell-free ascitic fluid aspirated, and the pellet resuspended in 1 ml of MEM, 20% NBS with 10 mg of bovine fibrinogen, and processed as described above. The viability of the implanted tumor was determined by mashing one of the fragments into trypan blue and estimating the proportion of viable cells. Ascites cell preparations were usually more than 80% viable. Chemotherapy. The tumor fragments were implanted on Day 0. The mice were administered the chemotherapy by intraperitoneal injections on Days 1 through 5. All test animals received combination chemotherapy of CAP: 0.08 mg of doxorubicin HCl (Adriamycin, Adria Laboratories, Inc., Columbus, Ohio), 0.04 mg of cis-diamminedichloroplatinum (Cisplatin, National Cancer Institute, Bethesda, Md.), and 1.0 mg of cyclophosphamide (Cytoxan, Mead Johnson & Co., Evansville, Ind.) each day. Control animals were injected with saline.
187
CHEMOSENSITIVITY TESTING OF NONSOLID TUMORS
RESULTS The growth of the human myeloid leukemia is illustrated in Table 1. At the lower concentrations of live tumor cells it was difficult to obtain a clot large and firm enough to cut into fragments. If the ratio of fibrinogen to cells was too high, the clot became sticky and was difficult to implant. There is a direct correlation between the growth of the implant and the number of live tumor cells present in the fragment. The growth of the tumor from three patients with adenocarcinoma of the ovary is illustrated in Table 2. In the cases in which we had both ascites cells and solid tumor, the ascites clot grew better than the solid tumor. However, this was probably due to the greatly increased viability of the tumor cells in the ascitic fluids. The sensitivity to CAP chemotherapy is the same. The third patient had numerous paracentesis. The first SRC implant assay was done with a natural clot from the fluid; it was not centrifuged. The second assay was prepared from a centrifuged and compacted clot. The third and fourth SRC-implant assays were done with fragments prepared from a centrifuged cell pellet with added fibrinogen to form a firm clot. All four of the tumor-cell preparations were sensitive to chemotherapy with CAP. This patient’s tumor also responded well to CAP chemotherapy clinically; however, she had received her lifetime dose of Adriamycin last year necessitating termination of therapy. Since that time she has developed increasing ascites. DISCUSSION The subrenal-capsule implant assay has been developed as an in vivo assay to predict the response of an individual patient’s tumor to chemotherapy. The assay site provides a rich vascular bed that permits unrestricted nutrient and drug access to the implanted tumor. Any metabolic steps required for drug activation are also provided by the animal. The assay results have been shown to have a very high correlation with the clinical tumor response in a wide variety of malignant tissues [I]. Our laboratory has an 84% positive correlation of chemotherapy sensitivity in this assay with the clinical response of the patient’s tumor to the same drugs [3]. Our negative correlations, lack of response in both the assay and clinically, are 99%. The one drawback of this assay has been the inability to test nonsolid tumors and malignant effusions. Many of the ovarian cancer patients develop malignant ascites and the ability to serially follow the GROWTH OF HUMAN
MYELOID
LEUKEMIA
TABLE 1 K562 IN SUBRENAL-CAPSULE
XENOGRAFT
ASSAY
No. live cells/implant
Implant size
Implant growth (AOMU)”
1 x lo6 4 x lo5 2 x lo5
8.6 ? 1.3 (9)b 8.0 I 1.3 (4) 7.5 f 0.9 (5)
8.4 2 4.9 (9) 2.3 2 0.2 (4) 2.1 f 2.2 (5)
a AOMU =
(final width + length) 2
-
(initial width + length)
2 b Mean * SD (number of animals) in OMU, where 10 OMU = 0.7 mm.
188
STRATTON
ET AL.
TABLE 2 THE GROWTH OF SOLID TUMOR TISSUE AND CLOTTED ASCITES TUMOR CELLS IN THE SUBRENALCAPSULE IMPLANT ASSAY: GROWTH AND SENSITIVITY TO COMBINATION CHEMOTHERAPY
Experiment No.
Tumor
Percentage viability
Control growth AOMU
Growth inhibition by CAP (% of control)
151
Solid Ascites
38 98
1.67 ? 0.93” 3.80 zt 2.20
- 120h -34
153
Solid Ascites
16 98
0.25 2 1.09 1.60 f 2.22
- 248 - 104
Ascites Ascites Ascites Ascites
98 80 98 99
1.30 2.50 7.20 2.10
-491 - 205 -123 -187
154
1 2 3 4
2 f 2 f
1.57 1.26 3.11 1.76
’ Mean 2 SD (final mean diameter of implant-initial mean diameter of implant). (AOMU control - AOMU test) x 1oo b Percentage of control = 1 AOMU Control
chemotherapy sensitivity of these malignant cells could significantly add to the effectiveness of the therapy provided to the patient. Thus, the modification of the assay to allow testing of nonsolid tumors would make the subrenal-capsule xenograft assay for chemotherapeutic sensitivity the assay of choice to select the appropriate drug with which to treat the patient. ACKNOWLEDGMENTS The authors wish to acknowledge the skilled technical assistance of Mr. Jack Shields and Ms. June Coverley for preparation of this manuscript.
REFERENCES 1. Bogden, A. E., Haskell, P. M., LePage, D. J., Kelton, D. E., Cobb, W. R., and Esber, H. J. Growth of human tumor xenografts implanted under the renal capsule of normal immunocompetent mice, Exp. Cell Biol. 47, 281-293 (1979). 2. Stratton, J. A., Rettenmaier, M. A., Braly, P. S., and DiSaia, P. J. Accurate predictions of tumor growth in viva: The subrenal capsule implant site, Proceedings 13th International Cancer Congress, Abstr. 1907, p. 335 (1982). 3. Braly, P. S., Stratton, J. A., and DiSaia, P. J. Preclinical assessments for chemotherapeutic responses of human gynecologic tumors, submitted for publication (1983).
ONCOLOGY
17, 185-188 (1984)
Chemosensitivity Testing of Nonsolid Tumors by the Subrenal-Capsule Implant Assay JOAN
A. STRATTON, Ph.D.,’ JOHN P. MICHA, M.D., MARK A. RETTENMAIER, M.D., PATRICIA S. BRALEY, M.D., AND PHILIP J. DISAIA, M.D.
University
of California,
Irvine Medical Center, Department of Obstetrics and Gynecology, IO1 City Drive, Orange, California 92668
Received August 18, 1982 A procedure to test chemosensitivity of nonsolid tumors using the subrenal-capsule xenograft assay has been developed. This technique is applicable to the study of malignant effusions and hematopoietic tumors. The tumor cells are centrifuged to form a pellet and the pellet is resuspended in the presence of bovine fibrinogen to form a jelled clot. This clot is cut into fragments which are implanted beneath the renal capsule of normal mice. The growth of the tumor cells from ovarian adenocarcinoma obtained from ascitic fluid is better than that of tumor fragments from the solid tumors. This is probably a reflection of the greater viability of tumor cells from ascitic fluids. The sensitivity to chemotherapy was the same for the solid tumor and its malignant effusions.
INTRODUCTION The subrenal-capsule assay method of determining chemosensitivity of human tumors was developed by Arthur Bogden and colleagues [l] at the Mason Research Institute in Worcester, Mass. This technique has the advantages of being an in vivo assay and any metabolic steps required for activation of chemotherapeutic drugs are performed by the intact animal. The assay provides employable information in a short period of time so that initiation of chemotherapy is not delayed. Since each tumor fragment serves as its own control, both oncolysis and oncostasis can be determined. And finally, combination chemotherapies can be assayed. We have found an 84% long-term positive correlation with the results of the assay and the response of patients with various gynecologic malignancies to chemotherapy [3]. This manuscript deals with solid tumors and the long-term (12 months) clinical follow-ups and validates the subrenal-capsule tumor implant assay as a preclinical predictor of clinical sensitivity to chemotherapeutic agents. To date only solid tumor explants have been studied in the subrenal-capsule xenograft assay. In this study, we report an adaptation of this assay which allows ’ To whom requests for reprints and correspondence should be addressed. 185 OWO-8258/84 $1.50 Copyright 0 1984 by Academic Press, Inc. All rights of reproduction in any form reserved.
186
STRATTON
ET AL.
chemosensitivity testing of malignant cell suspensions. Thus the chemosensitivity of the patient’s tumor may be followed without further surgery. MATERIALS AND METHODS Culture medium. Minimum Eagle’s medium (MEM, Grand Island Biological Co., Oakland, Calif.) containing 10,000 units each of penicillin and streptomycin (penicillin-streptomycin, GIBCO), 0.05 mg/ml gentamycin sulfate (Garamycin, Shering Corp., Kenilwood, N.J.), 2 mM glutamine (GIBCO), and 20% newborn calf serum (NBS, GIBCO). Single cell suspensions. The human myeloid leukemia cell line K562, obtained from S. H. Golub (Department of Surgery, University of California at Los Angeles) has been maintained in culture for several years. The viability (trypan-blue exclusion) and the number of tumor cells were determined by counting in a hemocytometer. Cell suspensions were prepared containing 1 x 107, 2 x 106, and 2 x lo5 live K.562 cells. The cell suspensions were centrifuged at 1200 rpm for 10 mitt, the culture medium was discarded, and the cell pellet was resuspended in 1 ml of MEM and 20% NBS. Ten milligrams of bovine fibrinogen (F-4000, Sigma Chemical Co., St. Louis, MO.) was added to the resuspended cells and the suspension centrifuged at 1000 rpm for 10 min. The pellet was incubated at 37°C for lo-30 min until a firm clot formed. The jelled cell pellet was poured into a cold petri dish with fresh cold MEM and 10% dextrose ((Y- D( +) glucose, Sigma), and then cut into small pieces (approximately 1 mm3). The fragments tend to be sticky; the dextrose helps to keep them separated. The fragments were implanted beneath the renal capsule of normal BDF, mice as described previously for solid tumors [ 1,2]. Malignant ascites was obtained by paracentesis from four patients with ovarian adenocarcinoma. Solid tumor was also obtained from two of these patients at laparotomy. The ascites was allowed to clot naturally, and the clot and sedimented cells were centrifuged at 1200 t-pm for 15 min. Firmer clots were obtained if the ascitic fluid was allowed to settle overnight at 4°C. If the natural clots were too friable, 10 mg of bovine fibrinogen was added to the vortexed clot, and the suspension was centrifuged and handled as described for the K562 leukemia. If no natural clot formed, the ascitic fluid was centrifuged at 1200 rpm for 15 mitt, the cell-free ascitic fluid aspirated, and the pellet resuspended in 1 ml of MEM, 20% NBS with 10 mg of bovine fibrinogen, and processed as described above. The viability of the implanted tumor was determined by mashing one of the fragments into trypan blue and estimating the proportion of viable cells. Ascites cell preparations were usually more than 80% viable. Chemotherapy. The tumor fragments were implanted on Day 0. The mice were administered the chemotherapy by intraperitoneal injections on Days 1 through 5. All test animals received combination chemotherapy of CAP: 0.08 mg of doxorubicin HCl (Adriamycin, Adria Laboratories, Inc., Columbus, Ohio), 0.04 mg of cis-diamminedichloroplatinum (Cisplatin, National Cancer Institute, Bethesda, Md.), and 1.0 mg of cyclophosphamide (Cytoxan, Mead Johnson & Co., Evansville, Ind.) each day. Control animals were injected with saline.
187
CHEMOSENSITIVITY TESTING OF NONSOLID TUMORS
RESULTS The growth of the human myeloid leukemia is illustrated in Table 1. At the lower concentrations of live tumor cells it was difficult to obtain a clot large and firm enough to cut into fragments. If the ratio of fibrinogen to cells was too high, the clot became sticky and was difficult to implant. There is a direct correlation between the growth of the implant and the number of live tumor cells present in the fragment. The growth of the tumor from three patients with adenocarcinoma of the ovary is illustrated in Table 2. In the cases in which we had both ascites cells and solid tumor, the ascites clot grew better than the solid tumor. However, this was probably due to the greatly increased viability of the tumor cells in the ascitic fluids. The sensitivity to CAP chemotherapy is the same. The third patient had numerous paracentesis. The first SRC implant assay was done with a natural clot from the fluid; it was not centrifuged. The second assay was prepared from a centrifuged and compacted clot. The third and fourth SRC-implant assays were done with fragments prepared from a centrifuged cell pellet with added fibrinogen to form a firm clot. All four of the tumor-cell preparations were sensitive to chemotherapy with CAP. This patient’s tumor also responded well to CAP chemotherapy clinically; however, she had received her lifetime dose of Adriamycin last year necessitating termination of therapy. Since that time she has developed increasing ascites. DISCUSSION The subrenal-capsule implant assay has been developed as an in vivo assay to predict the response of an individual patient’s tumor to chemotherapy. The assay site provides a rich vascular bed that permits unrestricted nutrient and drug access to the implanted tumor. Any metabolic steps required for drug activation are also provided by the animal. The assay results have been shown to have a very high correlation with the clinical tumor response in a wide variety of malignant tissues [I]. Our laboratory has an 84% positive correlation of chemotherapy sensitivity in this assay with the clinical response of the patient’s tumor to the same drugs [3]. Our negative correlations, lack of response in both the assay and clinically, are 99%. The one drawback of this assay has been the inability to test nonsolid tumors and malignant effusions. Many of the ovarian cancer patients develop malignant ascites and the ability to serially follow the GROWTH OF HUMAN
MYELOID
LEUKEMIA
TABLE 1 K562 IN SUBRENAL-CAPSULE
XENOGRAFT
ASSAY
No. live cells/implant
Implant size
Implant growth (AOMU)”
1 x lo6 4 x lo5 2 x lo5
8.6 ? 1.3 (9)b 8.0 I 1.3 (4) 7.5 f 0.9 (5)
8.4 2 4.9 (9) 2.3 2 0.2 (4) 2.1 f 2.2 (5)
a AOMU =
(final width + length) 2
-
(initial width + length)
2 b Mean * SD (number of animals) in OMU, where 10 OMU = 0.7 mm.
188
STRATTON
ET AL.
TABLE 2 THE GROWTH OF SOLID TUMOR TISSUE AND CLOTTED ASCITES TUMOR CELLS IN THE SUBRENALCAPSULE IMPLANT ASSAY: GROWTH AND SENSITIVITY TO COMBINATION CHEMOTHERAPY
Experiment No.
Tumor
Percentage viability
Control growth AOMU
Growth inhibition by CAP (% of control)
151
Solid Ascites
38 98
1.67 ? 0.93” 3.80 zt 2.20
- 120h -34
153
Solid Ascites
16 98
0.25 2 1.09 1.60 f 2.22
- 248 - 104
Ascites Ascites Ascites Ascites
98 80 98 99
1.30 2.50 7.20 2.10
-491 - 205 -123 -187
154
1 2 3 4
2 f 2 f
1.57 1.26 3.11 1.76
’ Mean 2 SD (final mean diameter of implant-initial mean diameter of implant). (AOMU control - AOMU test) x 1oo b Percentage of control = 1 AOMU Control
chemotherapy sensitivity of these malignant cells could significantly add to the effectiveness of the therapy provided to the patient. Thus, the modification of the assay to allow testing of nonsolid tumors would make the subrenal-capsule xenograft assay for chemotherapeutic sensitivity the assay of choice to select the appropriate drug with which to treat the patient. ACKNOWLEDGMENTS The authors wish to acknowledge the skilled technical assistance of Mr. Jack Shields and Ms. June Coverley for preparation of this manuscript.
REFERENCES 1. Bogden, A. E., Haskell, P. M., LePage, D. J., Kelton, D. E., Cobb, W. R., and Esber, H. J. Growth of human tumor xenografts implanted under the renal capsule of normal immunocompetent mice, Exp. Cell Biol. 47, 281-293 (1979). 2. Stratton, J. A., Rettenmaier, M. A., Braly, P. S., and DiSaia, P. J. Accurate predictions of tumor growth in viva: The subrenal capsule implant site, Proceedings 13th International Cancer Congress, Abstr. 1907, p. 335 (1982). 3. Braly, P. S., Stratton, J. A., and DiSaia, P. J. Preclinical assessments for chemotherapeutic responses of human gynecologic tumors, submitted for publication (1983).