- Email: [email protected]
A randomized comparison of a rapid versus prolonged (24 hr) infusion of cisplatin in therapy of squamous cell carcinoma of the uterine cervix: A gynecologic oncology group study
GYNECOLOGICONCOLOGY32, 198--202 (1989)
A Randomized Comparison of a Rapid versus Prolonged (24 hr) Infusion of Cisplatin in Therapy of Squamous Cell Carcinoma of the Uterine Cervix: A Gynecologic Oncology Group Study J. TATE THIGPEN, M.D., *'l JOHN A . BLESSING, PH.D.,? PHILIP J. DISAIA, M.D.,$ WESLEY C. FOWLER, JR., M.D.,§ AND KENNETH D. HATCH, M.D.]]
*Departments of Medicine and Obstetrics and Gynecology, University of Mississippi School of Medicine, Jackson, Mississippi; ?Gynecologic Oncology Group, Roswell Park Memorial Institute, Buffalo, New York; ¢cDepartment of Obstetrics and Gynecology, Director, University of California, Irvine Cancer Center, Orange, California; §Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of North Carolina School of Medicine, Chapel Hill, North Carolina; and UDivision of Gynecologic Oncology, The University of Alabama in Birmingham, Birmingham, Alabama Received June 28, 1988
In this study, 331 patients with advanced or recurrent squamous cell carcinoma of the cervix no longer amenable to control with surgery or radiotherapy were randomized to receive cisplatin 50 mg/m 2 as either a continuous infusion over 24 hr or a more rapid infusion at a rate of 1 mg/min. Antiemetic therapy was standardized for the initial course of both regimens as metoclopramide 60 mg at the time of and at 3 and 6 hours after initiation of cisplatin. The overall frequency of objective regression of disease was 18%; the response rate in each regimen was essentially identical. The continuous infusion regimen was associated with a significantly greater percentage of patients who experienced no nausea and vomiting (34% versus 18%, P = 0.002). Other adverse effects included nephrotoxicity, peripheral neuropathy, myelosuppression, and ototoxicity. Both the frequency and severity of these were essentially the same for each regimen. ©1989AcademicPress,Inc. INTRODUCTION The Gynecologic Oncology Group (GOG) discovered the activity of cisplatin against squamous cell carcinoma of the uterine cervix in the course of a series of phase II trials of single agents in patients who had received no prior chemotherapy [1]. The initial phase II observation that cisplatin is active against squamous cell carcinoma of the cervix has since been confirmed and a modest enhancement of activity documented by an increase in dose intensity [2]. Interest subsequently developed in the use of a continuous infusion schedule for the purpose of amelioration of nausea and vomiting as well as potential J To whom reprint requests should be addressed at GOG Headquarters, Suite 1945, 1234 Market Street, Philadelphia, PA 19107. 198 0090-8258/89 $1.50 Copyright © 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
enhancement of therapeutic effect. This latter study, GOG Protocol 64, is the subject of the current report. MATERIALS AND METHODS Between April 1982 and July 1985, 380 patients with biopsy-proven squamous cell carcinoma of the uterine cervix were entered into study by 30 member institutions of the GOG. The study required that the patient have advanced or recurrent disease no longer amenable to control with surgery a n d / o r radiotherapy and that lesions were measurable by physical examination or radiographic procedures. Additional eligibility criteria included a GOG performance status [3] of 0-3 (Karnofsky 30% or better), WBC 1> 3000/mcl, platelets ~>150,000/mcl, normal renal and hepatic function, no prior chemotherapy, recovery from the effects of prior radiotherapy or surgery, an absence of clinically significant infection, and no prior history of other malignancy. The patient was also required to have signed an informed consent. Among the 380 study entries, 43 were ruled ineligible and 6 inevaluable. The reasons for ineligibility included primary other than cervix, 7 patients; a second malignancy, 4 patients; cell type other than squamous cell carcinoma, 26 patients; no documentation of cervix primary, 4 patients; elevated blood urea nitrogen at entry, 1 patient; and GOG performance status of 4 at entry, 1 patient. All pathology was reviewed by a panel of the GOG Pathology Committee. Reasons for inevaluability included clerical error at entry, 1 patient; never treated, 1 patient; inadequate pathology materials, 1 patient; and inadequate data submitted, 2 patients. The distribution of these pa-
199
RAPID vs PROLONGED INFUSION OF CISPLATIN
tients was equal between the treatment regimens with 19 in one, 24 in the other. The 331 evaluable patients were randomized to receive either cisplatin 50 mg/m 2 intravenously at a rate of 1 mg/min repeated every 3 weeks, or the same drug and dose given as a continuous intravenous infusion over a 24-hr period every 3 weeks. Therapy was continued until disease progressed or adverse effects dictated cessation of therapy. Therapeutic efficacy was evaluated in terms of response, duration of response, and survival. Response was defined in standard GOG terminology. A complete response was disappearance of all gross disease, including disappearance of ascites and pleural effusion, for at least 1 month. A partial response was a 50% or greater decrease in the product of perpendicular diameters for each lesion for at least 1 month. Progressive disease was a 50% or greater increase in the product of perpendicular diameters of any lesion documented in two separate examinations at least 2 weeks apart or the appearance of any new lesion within 1 month. Stable disease was disease with less than a 50% change as described above. Duration of response was defined as the time from first documentation of response to date of definite progression or of last contact. Survival was the time from study entry to death or date of last contact. Adverse effects were generally graded as absent, mild, moderate, or severe according to GOG adverse effects criteria [3]. Particular attention was devoted to the evaluation of nausea and vomiting. Patients were assigned a point score according to the severity of nausea and vomiting over the 48 hr following initiation of each course of therapy. One episode of vomiting was assigned a score of 3 points, an episode of retching 2 points, and an episode of nausea 1 point. The first 24 hr were monitored by an oncology nurse, the second 24 hr by the patient. Antiemetics were limited to metoclopramide 60 mg intravenously prior to and every 3 hours following initiation of therapy for a total of three doses per course at least for the first course of cisplatin. The 48-hr point total on the first course of therapy was used as a basis of comparison. A patient was judged to have had no nausea and vomiting if her point score was 0. A point score of 1 to 10 was equated to mild effects, 11 to 20 moderate, 21 to 40 severe, and greater than 40 very severe. Scores were recorded on a special form (GOG L-l) (Fig. 1). Randomization of patients with equal probability into each of the two treatment regimens was carried out by employing a Latin Square arrangement and balancing the sequence of assigned regimens both within and across institutions. Comparisons between regimens with regard to response and degree of nausea and vomiting were conducted with a two-sided test. The probability of a type I error was fixed at 0.05, and the resultant power
was 0.80. In both comparisons, the study was designed to detect a 15% difference between regimens. RESULTS
Patient Characteristics The significant characteristics of the 331 evaluable patients are summarized in Table 1. Of the 331 patients, 168 were randomized to the rapid-infusion regimen, 163 to the continuous-infusion regimen. The median age in each group was 51 years. Over 70% of the cases in each regimen had a performance status of 0 or 1. Almost 90% of the population had received radiotherapy prior to study entry. In each regimen, a median of four courses of therapy was given. There were no major imbalances between the two regimens of the study withregard to these factors.
Response The overall response rate of those patients evaluable for response was 18% (Table 2). Nineteen of 320 (6%) patients achieved a complete response, 37 (12%) a partial response, 182 (57%) stable disease, and 82 (25%) progressive disease. There were no significant differences between the two treatment regimens. The median duration of complete responses was 7.4 months (range: 2-34), and the median survival for complete responders was 21.4 months (range: 6-40). The median duration of partial responses was 3.9 months (range: 1-24) with a median survival of those patients of 12.8 months (range: 6-45). The median survival for patients with stable disease was 7.1 months (range: 2-46), for those with progressive disease 3.0 months (range: 1-13). For the rapid infusion regimen, the median response duration was 4.5 months compared to 5.5 months for the continuous infusion regimen. There were no therapeutic differences in progression-free interval (PFI) (Fig. 2) or survival (Fig. 3). For the rapid infusion regimen median duration of PFI was 2.9 months and median survival was 6.2 months, while the corresponding values for the continuous infusion regimen were 6.4 and 3.5 months, respectively. Again, no differences were noted between the treatment regimens.
Adverse Effects Overall, 297 patients had data available showing the frequency and severity of nausea and vomiting for the first course of therapy (Table 3). Only 26% of those experienced no nausea and vomiting. Most had point scores between 1 and 40 with only 7% having very severe problems. There were significantly more patients who experienced no nausea and vomiting on the continuous infusion regimen (34% versus 18%, P = 0.002). Other adverse effects observed included peripheral
200
THIGPEN ET AL.
GYNECOLOGICONCOLOGYGROUP [( 7 i .... '77511
ANTIEMETIC REPORTING FORM
FO,,~..~
i I1' I-T"I'T°;I'"-='I "'"" ..... ....1 .....................................
REGIMEN (DRUG, NUMBER, AND TIME OF DOSES~
ANTIEMETIC
REPORTING PERIOD
VOMITING SUB EPISODESPOINTS TOTAL
RETCHIN~ SUB EPISODES POINTS TOTAL"
I
- -
X
3
=
- -
- -
X
2
=
- -
II
- -
X
3
=
- -
- -
X
2
=
- -
Ill
- -
X
3
=
- -
i
X
2
=
- -
TOTAL
POINTS
TOTAL m
m
D
DURING H O S P I T A L I Z A T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
lV
X
3
=
X
2
=
V
X
3
:
X
2
:
VI
X
3
=
X
2
:
TOTAL
NAUSEA
AFTER
OVERALL
THERAPY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
TOTAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
(
)
(
)
(
)
COMMENTS:
COPY 1 - MAIL TO G O G
OPERATIONS
OFFICE
FIGURE 1
neuropathy, azotemia, myelosuppression, and ototoxicity. Only 4% experienced neuropathy with only three cases exhibiting severe or life-threatening effects. While 49% of patients demonstrated azotemia, only six cases were TABLE 1
Relevant Characteristics of the Patient Population Patient characteristic Age Median Range
Rapid infusion
Continuous infusion TABLE 2
51 years 25-79 years
Response by Regimen
51 years 27-81 years
Performance status 0-1 2-3 Prior radiotherapy Number of courses Median Range
rated as severe, none as life-threatening. N o cases of life-threatening myelosuppression were noted, and only six patients experienced severe hematologic toxicity. Finally, three patients reported ototoxicity, only one severe. In no instance was an adverse effect noted to be more frequent or more severe in the continuous-infusion regimen. The only significant difference between the two
Response category 71% 29% 89%
77% 23% 90%
4 1-13
4 1-18
Complete response Partial response Stable disease Progressive disease Total
Rapid infusion (%) 10 18 90 46
(6%) (11%)
(55%) (28%)
164 (100%)
Continuous infusion (%) 9 19 92 36
(6%) (12%) (59%) (23%)
156 (100%)
Total (%) 19 37 182 82
(6%) (12%) (57%) (26%)
320 (100%)
201
RAPID vs PROLONGED INFUSION OF CISPLATIN
TABLE 3
Duration of Progression-Free Inferval By Regimen
Nausea and Vomiting by Regimen~
100 9O
I c .9 ~
,o
Nausea and vomiting
Continuous infusion
711 60-
5o-
'\ \
Regimen
'~
~Rapid
Progression Free Failures Total Median 9 156 165 2.9 151 159 3.5
None (score 0) Mild (score 1-10) Moderate (score 11-20) Severe (score 21-40) Very severe (score >40)
Q_ 4oao-
L 20-
g_
Rapid infusion
10" 0'0
1'2
tl
Months From Onset of Study
2~
3~
26 56 24 30 10
(17.8%) (38.4%) (16.4%) (20.6%) (6.8%)
51 51 13 25 11
(33.8%) (33.8%) (8.6%) (16.6%) (7.2%)
A significantly greater percentage (P = 0.002) receiving the continuous infusion experienced no nausea and vomiting as compared to patients receiving rapid infusion.
FIGURE 2
The principal problem noted in this and the previous patient population was nausea and vomiting which ocregimens was the difference noted with regard to nausea curred in 90% of cases. and vomiting. The desire to reduce the emetogenic potential of cisplatin and the hope of enhancing the therapeutic efficacy DISCUSSION prompted the GOG to initiate this randomized study of As of 1976, several cytotoxic drugs had been identified a 24-hour continuous intravenous infusion of cisplatin as having modest activity in the treatment of squamous versus the standard infusion at a rate of 1 mg/min. Certain cell carcinoma [4]. None were highly active, however, reports had suggested that such an approach might produce and attempts at combination chemotherapy were plentiful less nausea and vomiting. Investigators using a 24-hr but uncontrolled and often contradictory [5]. The GOG continuous intravenous infusion of cisplatin 80 mg/m 2 began at that time a series of phase II trials in patients every 3 weeks noted an apparent reduction in the severity with squamous cell carcinoma of the cervix in an effort and frequency of nausea and vomiting in a series of 18 to identify highly active agents which might logically be patients with advanced head and neck cancer [6]. Another combined. The second drug in that series, cisplatin, pro- series of 34 patients was given a 5-day infusion of 20 duced a response rate of 50% (11 responses among 22 mg/m2/day with a similar observation of apparently repatients) [1]. More importantly, cisplatin produced com- duced nausea and vomiting [7]. In neither of these trials, plete responses (3 among the 22 patients) and responses however, was antiemetic therapy controlled; and neither study had a randomized control group. The method of in disease confined to an irradiated pelvis. The dose of cisplatin in the initial trial was a modest assessment of nausea and vomiting was, furthermore, 50 mg/m 2. A second study of cisplatin demonstrated drug not formalized. The current study attempted to deal with the problems activity with three different dose schedules but failed to of these earlier investigations. A prospective, randomized show more than a modest increase in response rate [2] with an increase in dose. The overall response rate in control was employed; and antiemetic therapy was limited this trial was 26% (115 responses among 444 patients). to a standard dose of metoclopramide which was the most effective regimen available at the time of initiation of this trial [8]. The method of assessment chosen had Survival Probabilify been developed by the Nursing Committee of the GOG By Regimen i00-[~,. and used in a limited trial [9]. The results confirm the 90 ". earlier observations that, indeed, the prolonged infusion results in a reduction of nausea and vomiting. At least two reports suggested that a prolonged infusion 60 ,, Ra d 13 152 165 6.2 might have a therapeutic advantage. The killing effect of cisplatin against cultured human lymphoma cells was :~ 40 ",.,. 30 "-'-. enhanced by prolonged infusion as compared to pulse administration [10]. Studies of the pharmacokinetics and protein binding of cisplatin also pointed to a potential 0 6 12 18 24 SO advantage for continuous infusion [11]. The current trial Months From Onset of Study was designed to identify a 15% difference in response rate favoring either regimen. No such difference was FIGURE 3
202
THIGPEN ET AL.
found and, in fact, the response rates for each regimen are ~es~Ylal!y jdentica! ......... o~. . . . . . . . ~ ~, .........~.......
~r~Sbrted in the prevmus stud~, [2] but not slgmficantly ~ l~er.
In each of the GOG trmls mvolwng clsplatm m
responses among 791 patients). Changes in dose level have not resulted in marked changes in response rate. Change in the dose schedule, however, has resulted in a diminution of nausea and vomiting with prolongation of the time of infusion and a maintenance of therapeutic efficacy. In conclusion, the 24-hr continuous infusion of cisplatin offers an advantage over more rapid infusions in terms of reduced gastrointestinal toxicity. This improvement in the therapeutic index would seem to dictate the use of the prolonged infusion in those patients who experience cisplatin-related nausea and vomiting.
ACKNOWLEDGMENTS The following are participating institutions and the National Cancer Institute grants supporting this study. University of Alabama at Birmingham (CA 12484), The Oregon Health Sciences Center University, 2 Duke University Medical Center (CA 12534), Hahnemann Medical College/Jefferson Medical College/Temple University School of Medicine (CA 12478), University of Rochester Medical Center (CA 12482), Walter Reed Army Medical Center (CA 23501), University of Minnesota Medical School (CA 23088), University of Southern California Medical Center at Los Angeles (CA 37535), University of Mississippi Medical Center (CA 13633), Colorado Foundation for Medical Care (CA 15975), University of California Medical Center at Los Angeles (CA 13630), University of Miami School of Medicine (CA 37234), The Milton S. Hershey School of Medicine of Pennsylvania State University (CA 16386), Georgetown University Hospital (CA 16938), University of North Carolina School of Medicine (CA 23073), University of Iowa Hospitals and Clinics (CA 195902), University of Texas Health Science Center at Dallas (CA 28160), Indiana
2 Unfunded.
University Medical Center (CA 21720), Bowman Gray School of Medicine of Wake Forest University (CA 21946), The Albany Medical College of Union University, 2 University of California Medical Center at Irvine (CA 23765), Tufts New England Medical Center (CA 37569), Illinois Cancer Council (CA 27806), University of Pittsburgh School of Medicine,2 St. Louis University Medical Center (CA 35571), Stanford University Medical Center, 2 State University of New York Downstate Medical Center (CA 34477), Latter Day Saints Hospital, 2 Eastern Virginia Medical School,2 University of New Mexico, 2 University of Connecticut Medical Center, z and University of Kentucky. 2
REFERENCES 1. Thigpen, T., Shingleton, H., Homesley, H., et al. cis-Platinum in treatment of advanced or recurrent squamous cell carcinoma of the cervix: A phase II study of the Gynecologic Oncology Group, Cancer 48, 899-903 (1981). 2. Bonomi, P., Blessing, J., Stehman, F., et al. Randomized trial of three cisplatin dose schedules in squamous cell carcinoma of the cervix (a Gynecologic Oncology Group Study), J. Clin. OncoL 3, 1079-1085 (1985). 3. Blessing, J. A. Design, analysis and interpretation of chemotherapy trials in gynecologic cancer, in Chemotherapy o f gynecologic cancer, (G. Deppe, Ed.), A. R. Liss, New York, pp. 49-83 (1984). 4. DeVita, V., Jr., Wasserman, T., Young, R., et al. Perspectives and research in gynecologic oncology, Cancer 38, 509 (1976). 5. Thigpen, T., Vance, R., Balducci, L., and Blessing, J. Chemotherapy in the management of advanced or recurrent cervical and endometrial carcinoma, Cancer 48, 658 (1981). 6. Jacobs, C., Bertino, J., Goffinet, D., et al. 24-hour infusion of cisplatinum in head and neck cancers, Cancer 42, 2135-2140 (1978). 7. Salem, P., Hall, S., Benjamin, R., et al. Clinical phase I-II study of cis-dichloro-diammineplatinum (II) given by continuous IV infusion, Cancer Treat. Rep. 62, 1553-1555 (1978). 8. Gralla, R., Itri, L., Pisko, S., et al. Antiemetic efficacy of highdose metoclopramide: Randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting, N. Engl. J. Med. 305, 905-909 (1981). 9. Lewis, G. Personal communication. 10. Drewinko, B., Brown, B., and Gottlieb, J. The effect of cis-diamminedichloroplatinum (II) on cultured human lymphoma cells and its therapeutic implications, Cancer Res. 33, 3091-3095 (1973). 11. Gullo, J., Litterst, C. L., Macguire, P. J., et al. Pharmacokinetics and protein binding of cis-dichlorodiammineplatinum (II) administered as a one-hour or as a 20-hour infusion, Cancer Pharmacol. 5, 21-26 (1980).
A Randomized Comparison of a Rapid versus Prolonged (24 hr) Infusion of Cisplatin in Therapy of Squamous Cell Carcinoma of the Uterine Cervix: A Gynecologic Oncology Group Study J. TATE THIGPEN, M.D., *'l JOHN A . BLESSING, PH.D.,? PHILIP J. DISAIA, M.D.,$ WESLEY C. FOWLER, JR., M.D.,§ AND KENNETH D. HATCH, M.D.]]
*Departments of Medicine and Obstetrics and Gynecology, University of Mississippi School of Medicine, Jackson, Mississippi; ?Gynecologic Oncology Group, Roswell Park Memorial Institute, Buffalo, New York; ¢cDepartment of Obstetrics and Gynecology, Director, University of California, Irvine Cancer Center, Orange, California; §Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of North Carolina School of Medicine, Chapel Hill, North Carolina; and UDivision of Gynecologic Oncology, The University of Alabama in Birmingham, Birmingham, Alabama Received June 28, 1988
In this study, 331 patients with advanced or recurrent squamous cell carcinoma of the cervix no longer amenable to control with surgery or radiotherapy were randomized to receive cisplatin 50 mg/m 2 as either a continuous infusion over 24 hr or a more rapid infusion at a rate of 1 mg/min. Antiemetic therapy was standardized for the initial course of both regimens as metoclopramide 60 mg at the time of and at 3 and 6 hours after initiation of cisplatin. The overall frequency of objective regression of disease was 18%; the response rate in each regimen was essentially identical. The continuous infusion regimen was associated with a significantly greater percentage of patients who experienced no nausea and vomiting (34% versus 18%, P = 0.002). Other adverse effects included nephrotoxicity, peripheral neuropathy, myelosuppression, and ototoxicity. Both the frequency and severity of these were essentially the same for each regimen. ©1989AcademicPress,Inc. INTRODUCTION The Gynecologic Oncology Group (GOG) discovered the activity of cisplatin against squamous cell carcinoma of the uterine cervix in the course of a series of phase II trials of single agents in patients who had received no prior chemotherapy [1]. The initial phase II observation that cisplatin is active against squamous cell carcinoma of the cervix has since been confirmed and a modest enhancement of activity documented by an increase in dose intensity [2]. Interest subsequently developed in the use of a continuous infusion schedule for the purpose of amelioration of nausea and vomiting as well as potential J To whom reprint requests should be addressed at GOG Headquarters, Suite 1945, 1234 Market Street, Philadelphia, PA 19107. 198 0090-8258/89 $1.50 Copyright © 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
enhancement of therapeutic effect. This latter study, GOG Protocol 64, is the subject of the current report. MATERIALS AND METHODS Between April 1982 and July 1985, 380 patients with biopsy-proven squamous cell carcinoma of the uterine cervix were entered into study by 30 member institutions of the GOG. The study required that the patient have advanced or recurrent disease no longer amenable to control with surgery a n d / o r radiotherapy and that lesions were measurable by physical examination or radiographic procedures. Additional eligibility criteria included a GOG performance status [3] of 0-3 (Karnofsky 30% or better), WBC 1> 3000/mcl, platelets ~>150,000/mcl, normal renal and hepatic function, no prior chemotherapy, recovery from the effects of prior radiotherapy or surgery, an absence of clinically significant infection, and no prior history of other malignancy. The patient was also required to have signed an informed consent. Among the 380 study entries, 43 were ruled ineligible and 6 inevaluable. The reasons for ineligibility included primary other than cervix, 7 patients; a second malignancy, 4 patients; cell type other than squamous cell carcinoma, 26 patients; no documentation of cervix primary, 4 patients; elevated blood urea nitrogen at entry, 1 patient; and GOG performance status of 4 at entry, 1 patient. All pathology was reviewed by a panel of the GOG Pathology Committee. Reasons for inevaluability included clerical error at entry, 1 patient; never treated, 1 patient; inadequate pathology materials, 1 patient; and inadequate data submitted, 2 patients. The distribution of these pa-
199
RAPID vs PROLONGED INFUSION OF CISPLATIN
tients was equal between the treatment regimens with 19 in one, 24 in the other. The 331 evaluable patients were randomized to receive either cisplatin 50 mg/m 2 intravenously at a rate of 1 mg/min repeated every 3 weeks, or the same drug and dose given as a continuous intravenous infusion over a 24-hr period every 3 weeks. Therapy was continued until disease progressed or adverse effects dictated cessation of therapy. Therapeutic efficacy was evaluated in terms of response, duration of response, and survival. Response was defined in standard GOG terminology. A complete response was disappearance of all gross disease, including disappearance of ascites and pleural effusion, for at least 1 month. A partial response was a 50% or greater decrease in the product of perpendicular diameters for each lesion for at least 1 month. Progressive disease was a 50% or greater increase in the product of perpendicular diameters of any lesion documented in two separate examinations at least 2 weeks apart or the appearance of any new lesion within 1 month. Stable disease was disease with less than a 50% change as described above. Duration of response was defined as the time from first documentation of response to date of definite progression or of last contact. Survival was the time from study entry to death or date of last contact. Adverse effects were generally graded as absent, mild, moderate, or severe according to GOG adverse effects criteria [3]. Particular attention was devoted to the evaluation of nausea and vomiting. Patients were assigned a point score according to the severity of nausea and vomiting over the 48 hr following initiation of each course of therapy. One episode of vomiting was assigned a score of 3 points, an episode of retching 2 points, and an episode of nausea 1 point. The first 24 hr were monitored by an oncology nurse, the second 24 hr by the patient. Antiemetics were limited to metoclopramide 60 mg intravenously prior to and every 3 hours following initiation of therapy for a total of three doses per course at least for the first course of cisplatin. The 48-hr point total on the first course of therapy was used as a basis of comparison. A patient was judged to have had no nausea and vomiting if her point score was 0. A point score of 1 to 10 was equated to mild effects, 11 to 20 moderate, 21 to 40 severe, and greater than 40 very severe. Scores were recorded on a special form (GOG L-l) (Fig. 1). Randomization of patients with equal probability into each of the two treatment regimens was carried out by employing a Latin Square arrangement and balancing the sequence of assigned regimens both within and across institutions. Comparisons between regimens with regard to response and degree of nausea and vomiting were conducted with a two-sided test. The probability of a type I error was fixed at 0.05, and the resultant power
was 0.80. In both comparisons, the study was designed to detect a 15% difference between regimens. RESULTS
Patient Characteristics The significant characteristics of the 331 evaluable patients are summarized in Table 1. Of the 331 patients, 168 were randomized to the rapid-infusion regimen, 163 to the continuous-infusion regimen. The median age in each group was 51 years. Over 70% of the cases in each regimen had a performance status of 0 or 1. Almost 90% of the population had received radiotherapy prior to study entry. In each regimen, a median of four courses of therapy was given. There were no major imbalances between the two regimens of the study withregard to these factors.
Response The overall response rate of those patients evaluable for response was 18% (Table 2). Nineteen of 320 (6%) patients achieved a complete response, 37 (12%) a partial response, 182 (57%) stable disease, and 82 (25%) progressive disease. There were no significant differences between the two treatment regimens. The median duration of complete responses was 7.4 months (range: 2-34), and the median survival for complete responders was 21.4 months (range: 6-40). The median duration of partial responses was 3.9 months (range: 1-24) with a median survival of those patients of 12.8 months (range: 6-45). The median survival for patients with stable disease was 7.1 months (range: 2-46), for those with progressive disease 3.0 months (range: 1-13). For the rapid infusion regimen, the median response duration was 4.5 months compared to 5.5 months for the continuous infusion regimen. There were no therapeutic differences in progression-free interval (PFI) (Fig. 2) or survival (Fig. 3). For the rapid infusion regimen median duration of PFI was 2.9 months and median survival was 6.2 months, while the corresponding values for the continuous infusion regimen were 6.4 and 3.5 months, respectively. Again, no differences were noted between the treatment regimens.
Adverse Effects Overall, 297 patients had data available showing the frequency and severity of nausea and vomiting for the first course of therapy (Table 3). Only 26% of those experienced no nausea and vomiting. Most had point scores between 1 and 40 with only 7% having very severe problems. There were significantly more patients who experienced no nausea and vomiting on the continuous infusion regimen (34% versus 18%, P = 0.002). Other adverse effects observed included peripheral
200
THIGPEN ET AL.
GYNECOLOGICONCOLOGYGROUP [( 7 i .... '77511
ANTIEMETIC REPORTING FORM
FO,,~..~
i I1' I-T"I'T°;I'"-='I "'"" ..... ....1 .....................................
REGIMEN (DRUG, NUMBER, AND TIME OF DOSES~
ANTIEMETIC
REPORTING PERIOD
VOMITING SUB EPISODESPOINTS TOTAL
RETCHIN~ SUB EPISODES POINTS TOTAL"
I
- -
X
3
=
- -
- -
X
2
=
- -
II
- -
X
3
=
- -
- -
X
2
=
- -
Ill
- -
X
3
=
- -
i
X
2
=
- -
TOTAL
POINTS
TOTAL m
m
D
DURING H O S P I T A L I Z A T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
lV
X
3
=
X
2
=
V
X
3
:
X
2
:
VI
X
3
=
X
2
:
TOTAL
NAUSEA
AFTER
OVERALL
THERAPY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
TOTAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
(
)
(
)
(
)
COMMENTS:
COPY 1 - MAIL TO G O G
OPERATIONS
OFFICE
FIGURE 1
neuropathy, azotemia, myelosuppression, and ototoxicity. Only 4% experienced neuropathy with only three cases exhibiting severe or life-threatening effects. While 49% of patients demonstrated azotemia, only six cases were TABLE 1
Relevant Characteristics of the Patient Population Patient characteristic Age Median Range
Rapid infusion
Continuous infusion TABLE 2
51 years 25-79 years
Response by Regimen
51 years 27-81 years
Performance status 0-1 2-3 Prior radiotherapy Number of courses Median Range
rated as severe, none as life-threatening. N o cases of life-threatening myelosuppression were noted, and only six patients experienced severe hematologic toxicity. Finally, three patients reported ototoxicity, only one severe. In no instance was an adverse effect noted to be more frequent or more severe in the continuous-infusion regimen. The only significant difference between the two
Response category 71% 29% 89%
77% 23% 90%
4 1-13
4 1-18
Complete response Partial response Stable disease Progressive disease Total
Rapid infusion (%) 10 18 90 46
(6%) (11%)
(55%) (28%)
164 (100%)
Continuous infusion (%) 9 19 92 36
(6%) (12%) (59%) (23%)
156 (100%)
Total (%) 19 37 182 82
(6%) (12%) (57%) (26%)
320 (100%)
201
RAPID vs PROLONGED INFUSION OF CISPLATIN
TABLE 3
Duration of Progression-Free Inferval By Regimen
Nausea and Vomiting by Regimen~
100 9O
I c .9 ~
,o
Nausea and vomiting
Continuous infusion
711 60-
5o-
'\ \
Regimen
'~
~Rapid
Progression Free Failures Total Median 9 156 165 2.9 151 159 3.5
None (score 0) Mild (score 1-10) Moderate (score 11-20) Severe (score 21-40) Very severe (score >40)
Q_ 4oao-
L 20-
g_
Rapid infusion
10" 0'0
1'2
tl
Months From Onset of Study
2~
3~
26 56 24 30 10
(17.8%) (38.4%) (16.4%) (20.6%) (6.8%)
51 51 13 25 11
(33.8%) (33.8%) (8.6%) (16.6%) (7.2%)
A significantly greater percentage (P = 0.002) receiving the continuous infusion experienced no nausea and vomiting as compared to patients receiving rapid infusion.
FIGURE 2
The principal problem noted in this and the previous patient population was nausea and vomiting which ocregimens was the difference noted with regard to nausea curred in 90% of cases. and vomiting. The desire to reduce the emetogenic potential of cisplatin and the hope of enhancing the therapeutic efficacy DISCUSSION prompted the GOG to initiate this randomized study of As of 1976, several cytotoxic drugs had been identified a 24-hour continuous intravenous infusion of cisplatin as having modest activity in the treatment of squamous versus the standard infusion at a rate of 1 mg/min. Certain cell carcinoma [4]. None were highly active, however, reports had suggested that such an approach might produce and attempts at combination chemotherapy were plentiful less nausea and vomiting. Investigators using a 24-hr but uncontrolled and often contradictory [5]. The GOG continuous intravenous infusion of cisplatin 80 mg/m 2 began at that time a series of phase II trials in patients every 3 weeks noted an apparent reduction in the severity with squamous cell carcinoma of the cervix in an effort and frequency of nausea and vomiting in a series of 18 to identify highly active agents which might logically be patients with advanced head and neck cancer [6]. Another combined. The second drug in that series, cisplatin, pro- series of 34 patients was given a 5-day infusion of 20 duced a response rate of 50% (11 responses among 22 mg/m2/day with a similar observation of apparently repatients) [1]. More importantly, cisplatin produced com- duced nausea and vomiting [7]. In neither of these trials, plete responses (3 among the 22 patients) and responses however, was antiemetic therapy controlled; and neither study had a randomized control group. The method of in disease confined to an irradiated pelvis. The dose of cisplatin in the initial trial was a modest assessment of nausea and vomiting was, furthermore, 50 mg/m 2. A second study of cisplatin demonstrated drug not formalized. The current study attempted to deal with the problems activity with three different dose schedules but failed to of these earlier investigations. A prospective, randomized show more than a modest increase in response rate [2] with an increase in dose. The overall response rate in control was employed; and antiemetic therapy was limited this trial was 26% (115 responses among 444 patients). to a standard dose of metoclopramide which was the most effective regimen available at the time of initiation of this trial [8]. The method of assessment chosen had Survival Probabilify been developed by the Nursing Committee of the GOG By Regimen i00-[~,. and used in a limited trial [9]. The results confirm the 90 ". earlier observations that, indeed, the prolonged infusion results in a reduction of nausea and vomiting. At least two reports suggested that a prolonged infusion 60 ,, Ra d 13 152 165 6.2 might have a therapeutic advantage. The killing effect of cisplatin against cultured human lymphoma cells was :~ 40 ",.,. 30 "-'-. enhanced by prolonged infusion as compared to pulse administration [10]. Studies of the pharmacokinetics and protein binding of cisplatin also pointed to a potential 0 6 12 18 24 SO advantage for continuous infusion [11]. The current trial Months From Onset of Study was designed to identify a 15% difference in response rate favoring either regimen. No such difference was FIGURE 3
202
THIGPEN ET AL.
found and, in fact, the response rates for each regimen are ~es~Ylal!y jdentica! ......... o~. . . . . . . . ~ ~, .........~.......
~r~Sbrted in the prevmus stud~, [2] but not slgmficantly ~ l~er.
In each of the GOG trmls mvolwng clsplatm m
responses among 791 patients). Changes in dose level have not resulted in marked changes in response rate. Change in the dose schedule, however, has resulted in a diminution of nausea and vomiting with prolongation of the time of infusion and a maintenance of therapeutic efficacy. In conclusion, the 24-hr continuous infusion of cisplatin offers an advantage over more rapid infusions in terms of reduced gastrointestinal toxicity. This improvement in the therapeutic index would seem to dictate the use of the prolonged infusion in those patients who experience cisplatin-related nausea and vomiting.
ACKNOWLEDGMENTS The following are participating institutions and the National Cancer Institute grants supporting this study. University of Alabama at Birmingham (CA 12484), The Oregon Health Sciences Center University, 2 Duke University Medical Center (CA 12534), Hahnemann Medical College/Jefferson Medical College/Temple University School of Medicine (CA 12478), University of Rochester Medical Center (CA 12482), Walter Reed Army Medical Center (CA 23501), University of Minnesota Medical School (CA 23088), University of Southern California Medical Center at Los Angeles (CA 37535), University of Mississippi Medical Center (CA 13633), Colorado Foundation for Medical Care (CA 15975), University of California Medical Center at Los Angeles (CA 13630), University of Miami School of Medicine (CA 37234), The Milton S. Hershey School of Medicine of Pennsylvania State University (CA 16386), Georgetown University Hospital (CA 16938), University of North Carolina School of Medicine (CA 23073), University of Iowa Hospitals and Clinics (CA 195902), University of Texas Health Science Center at Dallas (CA 28160), Indiana
2 Unfunded.
University Medical Center (CA 21720), Bowman Gray School of Medicine of Wake Forest University (CA 21946), The Albany Medical College of Union University, 2 University of California Medical Center at Irvine (CA 23765), Tufts New England Medical Center (CA 37569), Illinois Cancer Council (CA 27806), University of Pittsburgh School of Medicine,2 St. Louis University Medical Center (CA 35571), Stanford University Medical Center, 2 State University of New York Downstate Medical Center (CA 34477), Latter Day Saints Hospital, 2 Eastern Virginia Medical School,2 University of New Mexico, 2 University of Connecticut Medical Center, z and University of Kentucky. 2
REFERENCES 1. Thigpen, T., Shingleton, H., Homesley, H., et al. cis-Platinum in treatment of advanced or recurrent squamous cell carcinoma of the cervix: A phase II study of the Gynecologic Oncology Group, Cancer 48, 899-903 (1981). 2. Bonomi, P., Blessing, J., Stehman, F., et al. Randomized trial of three cisplatin dose schedules in squamous cell carcinoma of the cervix (a Gynecologic Oncology Group Study), J. Clin. OncoL 3, 1079-1085 (1985). 3. Blessing, J. A. Design, analysis and interpretation of chemotherapy trials in gynecologic cancer, in Chemotherapy o f gynecologic cancer, (G. Deppe, Ed.), A. R. Liss, New York, pp. 49-83 (1984). 4. DeVita, V., Jr., Wasserman, T., Young, R., et al. Perspectives and research in gynecologic oncology, Cancer 38, 509 (1976). 5. Thigpen, T., Vance, R., Balducci, L., and Blessing, J. Chemotherapy in the management of advanced or recurrent cervical and endometrial carcinoma, Cancer 48, 658 (1981). 6. Jacobs, C., Bertino, J., Goffinet, D., et al. 24-hour infusion of cisplatinum in head and neck cancers, Cancer 42, 2135-2140 (1978). 7. Salem, P., Hall, S., Benjamin, R., et al. Clinical phase I-II study of cis-dichloro-diammineplatinum (II) given by continuous IV infusion, Cancer Treat. Rep. 62, 1553-1555 (1978). 8. Gralla, R., Itri, L., Pisko, S., et al. Antiemetic efficacy of highdose metoclopramide: Randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting, N. Engl. J. Med. 305, 905-909 (1981). 9. Lewis, G. Personal communication. 10. Drewinko, B., Brown, B., and Gottlieb, J. The effect of cis-diamminedichloroplatinum (II) on cultured human lymphoma cells and its therapeutic implications, Cancer Res. 33, 3091-3095 (1973). 11. Gullo, J., Litterst, C. L., Macguire, P. J., et al. Pharmacokinetics and protein binding of cis-dichlorodiammineplatinum (II) administered as a one-hour or as a 20-hour infusion, Cancer Pharmacol. 5, 21-26 (1980).