Phase II Trial of Mitotane and Cisplatin in Patients With Adrenal Carcinoma: A Southwest Oncology Group Study
By Ronald M. Bukowski, Michael Wolfe, Howard S. Levine, David E. Crawford, Ronald L. Stephens, Ellen Gaynor, and W. Graydon Harker
Purpose: Previous reports of chemotherapy in pa- tients with adrenal cancer have described responses to cisplatin (CDDP). Because of these reports of good results, a phase II trial that used CDDP with and without mitotane (o,p’DDD) was initiated.
Patients and Methods: Patients with metastatic or residual adrenocortical carcinoma with objectively measurable disease or biochemical abnormalities were divided into good-risk and poor-risk categories. The lat- ter received CDDP 100 mg/m2 intravenously, and the former received 75 mg/m2. o,p’DDD was administered at a 1,000-mg dose orally four times a day along with cortisone acetate and Florinef (fludrocortisone acetate; Bristol-Myers Squibb Co, Princeton, NJ).
Results: Of a total of 42 patients entered onto the study, 37 were eligible. Twenty-nine patients received good-risk and eight received poor-risk doses of CDDP. Functioning tumors were present in 45% of patients.
A DRENAL CARCINOMA is a rare tumor with an estimated frequency of less than two cases per 1 million people in the United States.1,2 The natural history of this cancer is variable, but when metastatic disease de- velops, median survival periods ≤ 1 year have been re- ported.3.4 Patients with adrenocortical carcinoma often present with clinical syndromes secondary to overpro- duction of corticosteroids, and it has been estimated that 85% of these individuals will have evidence of functioning tumors.5
Therapy for this malignancy has been surgical, but dur- ing the past 30 years other approaches have been devel- oped. Since its clinical introduction in 1960, mitotane (o,p’DDD) has reportedly produced tumor regression in selected patients with metastatic disease. Initially, response rates of 30%2 to 60%6 were noted, but recent analyses have indicated that objective tumor regression is uncom- mon.7 Use of chemotherapy also has been investigated, but initial reports were anecdotal, and only small groups of patients were treated.7,8 Agents that produced responses included doxorubicin,7 carmustine (BCNU),9 fluoroura- cil,10 and methotrexate.11 Recently, Grapski et al12 and Tattersal et al13 reported that cisplatin (CDDP) produced objective tumor regressions in seven of eight patients treated. In view of the rarity of this tumor, and the pos- sibility that CDDP may have antitumor activity, a phase II trial using o,p’DDD in combination with CDDP was initiated in 1984. The purpose of the present report is to summarize the findings of this study.
Objective responses were noted in 30% (11 of 37) pa- tients (95% confidence interval, 16% to 50%). Response duration was 7.9 months, and the median time to re- sponse was 76 days. The median survival of the 37 eli- gible patients was 11.8 months, and a significant survival advantage was found for patients who underwent prior surgical removal of their primary tumor or bulky disease, who had a performance status of 0 or 1, or who had synchronous metastatic disease. Toxicity of the CDDP and o,p’DDD combination was moderate to severe, and the most common side effects were gastrointestinal, renal, and neurologic.
Conclusion: The regimen of CDDP and o,p’DDD has activity in patients with adrenocortical carcinoma; how- ever, the toxicity of this treatment was moderate to se- vere.
J Clin Oncol 11:161-165. @ 1993 by American Society of Clinical Oncology.
PATIENTS AND METHODS
Patients with metastatic or residual adrenocortical carcinoma in whom complete surgical removal of disease was not possible were eligible. Objectively measurable disease or the presence of a bio- chemical abnormality specific to a patient’s tumor were required. Prior therapy with o,p’DDD, chemotherapy (except CDDP) or ra- diotherapy was allowed. Hematologic and biochemical criteria in- cluded the following: WBC count ≥ 3,500/uL, platelet count ≥ 100,000/uL, serum creatinine level ≤ 1.5 mg/dL, and creatinine clearance ≥ 60 mL/min (measured or calculated). Pathology review was required, and failure to submit material for review rendered a patient ineligible. Finally, all patients were required to give written informed consent in accordance with institutional and Food and Drug Administration guidelines.
For the purposes of therapy, patients were divided into good-risk
From the Cleveland Clinic Foundation, Cleveland, OH; Southwest Oncology Group Statistical Center, Seattle, WA; University of Col- orado, Denver, CO; University of Kansas Medical Center, Kansas City, KS; Loyola University Stritch School of Medicine, Maywood, IL; and University of Utah Medical Center, Salt Lake City, UT. Submitted May 11, 1992; accepted August 10, 1992.
Supported in part by the following Public Health Service Cooperative Agreement grants awarded by the National Cancer Institute, De- partment of Health and Human Services: CA-04919, CA-37429, CA-42777, CA-12644, CA-46282, CA-13238, CA-16385, CA-04920, CA-45560, CA-20319, CA-35117, CA-52386, CA-22411, CA-35084, CA-46441, CA-03096, CA-35178, CA-35176, CA-36020, CA-46113, CA-35192, CA-14028, CA-35431, CA-12213, and CA-32102.
Address reprint requests to Southwest Oncology Group (SWOG- 8325), Operations Office, 5430 Fredericksburg Rd, Suite 168, San Antonio, TX 78229-6197.
C) 1993 by American Society of Clinical Oncology. 0732-183X/93/1101-0025$3.00/0
and poor-risk categories; the latter group received lower doses of CDDP. Poor-risk was defined as any one of the following present: age ≥ 65 years, extensive prior radiation ≥ 30% of the bone marrow- bearing areas, or poor tolerance to prior chemotherapy. CDDP and o,p’DDD were administered as listed in Table 1. All patients received concomitant cortisone acetate and Florinef (fludrocortisone acetate; Bristol-Myers Squibb Co, Princeton, NJ) during treatment with o,p’DDD. Hydration with 2,000 mL of 5% dextrose in half-normal saline in the 24 hours before administration of CDDP, and 1,000 mL over a 2-hour period after CDDP administration were required. A 12.5-g dose of mannitol before and 25 g after administration of CDDP also was given. CDDP was given every 3 weeks for 18 months or until progressive disease and/or unacceptable toxicity were ob- served.
Dose modification of CDDP and o,p’DDD was permitted and was based on toxicity. During treatment, a decrease in the creatinine clearance ≥ 30 mL/min and ≤ 50 mL/min required CDDP to be withheld until clearance normalized (≥ 50 mL/min). CDDP was discontinued if the creatinine clearance decreased to < 30 mL/min or did not normalize. In the presence of unacceptable gastrointestinal or neuromuscular side effects, o,p’DDD was decreased to 500 mg four times per day, and if after 14 days no improvement was noted, the dose was further decreased to 250 mg four times per day. If un- acceptable toxicity continued, o,p’DDD was then discontinued.
Studies performed before therapy included a complete history and physical examination, complete blood cell count with platelets, uri- nalysis, creatinine clearance, and determination of the levels of cre- atinine, lactic dehydrogenase, blood urea nitrogen (BUN), total bil- irubin, AST, calcium, phosphate, magnesium, and alkaline phosphatase. Twenty-four-hour urine collections for determination of 17-ketosteroids and 17-hydroxycorticosteroids were performed before therapy and also every 3 weeks if initially abnormal. Baseline chest radiographs and audiograms also were obtained and were re- peated as clinically indicated. X-ray studies or scans were performed every 3 weeks (6 weeks for computerized tomography) to determine the size of measurable lesions.
Response and Toxicity Criteria
A complete response was defined as the disappearance of all clinical evidence of tumor for a minimum of 4 weeks. Partial remission re- quired a ≥ 50% decrease in the sum of the products of the perpen- dicular diameters of measured lesions or a decrease ≥ 30% of the sum of liver measurements below the costal margins in the right, left, and xiphoid lines for a minimum of 4 weeks. Progressive disease was defined as a ≥ 25% increase in the size of any measured lesion or the appearance of new lesions. In patients without bidimensionally measurable disease, normalization of tumor products on two suc- cessive measurements was considered a complete response, and a 50% reduction was considered a partial response. Progression of bio- chemical parameters was present when a ≥ 50% increase in two con- secutive measurements was present. Toxicity was graded using Southwest Oncology Group (SWOG) criteria as previously de- scribed.14
Pathology Review
Pathologic material from 40 patients was reviewed. The number of slides received was recorded in 39 patients and ranged from one to 37 per patient, with a median of six and mean of 13. Tissue from the following sites was examined: the adrenal gland in 27 patients, the liver in nine, the lymph nodes in 14, the lung in three, the retro-
| Good-Risk | Poor-Risk* | |
|---|---|---|
| CDDP (every 3 weeks) o,p'DDD | 1,000 mg/m2 intravenously | 75 mg/m2 intravenously |
| 1,000 mg four times per day | 1,000 mg four times per day |
*Age ≥ 65 years, poor tolerance to prior chemotherapy or extensive prior radiotherapy.
peritoneum in three, the spleen in two, and the omentum, suprarenal area, adrenal bed, peritoneum, diaphragm, pleura, perinephric fat, kidney, and abdominal mass in one patient each. Of the 40 patients whose tissues were reviewed, one patient was ineligible because only an aspiration of a lung nodule was submitted and no malignant cells were present.
Biostatistical Methods
The primary objective of this study was to estimate the overall response rate. It was assumed that a true response rate of ≥ 20% would be of considerable interest in the treatment of metastatic or recurrent adrenal carcinoma.
An early-stop rule was applied separately to the subgroups of pre- viously treated versus untreated patients to rule out a 20% response rate with a 4.4% significance level; at least one response among the first 14 assessable patients was required to continue accrual in a subgroup.
A combined accrual goal of 28 assessable patients was selected to allow the overall response rate to be estimated with a standard de- viation of less than 10%.
RESULTS
The study was activated on January 19, 1984, and closed to patient accrual on July 1, 1990. Of the 42 patients entered onto the study, five were ineligible. Reasons for ineligibility included the following: no malignancy on pa- thology review (one patient); no pathology review (two patients); elevated pretreatment creatinine level (one pa- tient); and inadequate baseline date (one patient).
Pathologic review of 40 patients was performed, with one patient declared ineligible because malignancy could not be identified in the submitted material. Specimens were examined in two contiguous high-power microscopic fields for the presence of microscopic necrosis, calcifica- tion, mitoses, atypical mitoses, capsular invasion, and blood vessel invasion. Because the pathologic material submitted varied considerably with regard to amount of tissue, quality, and site sampled, the diagnosis of adrenal cortical carcinoma could not be rendered with absolute certainty on a histologic basis alone. In some cases the diagnosis was based on the histologic material in con- junction with clinical findings. Tissue from the adrenal tumor was available in 27 patients, with a sample from only a metastatic site in 10 instances. In one patient an adrenal carcinoma was present, as was a primary renal
cell carcinoma. This patient was considered eligible be- cause the entry criteria did not exclude patients with a second malignancy, and evidence of a functioning tumor was present. Data were insufficient to reach any conclu- sions with regard to correlations between microscopic features and response or survival.
Of the 37 eligible patients, 19 were male and 18 were female. Their median age was 54 years (range, 16 to 71). Twenty-nine patients received good-risk doses of CDDP, and eight received poor-risk doses. The majority of pa- tients (25 of 37, 68%) had not been previously treated and were ambulatory, with a performance status of 0 or 1 (27 of 37, 73%). Prior therapy in 12 patients included the following: o,p’DDD, six; radiation, five; chemotherapy, two; bacille Calmette-Guérin, one. The functional status of patients’ tumors was assessed before initiation of ther- apy, and clinical evidence of hyperfunctioning and/or el- evated urinary 17-hydroxycorticosteroids or 17-ketoste- roids was present in 15 of 33 (45%) patients in whom these values were reported. Only nine of 37 (24%) patients had not undergone previous adrenal surgery or surgery for removal of bulky abdominal disease.
Objective responses were reported in 11 of 37 eligible patients (one complete and 10 partial responses) for an overall response rate of 30% (95% confidence interval, 16.0% to 50%). The median response duration was 7.9 months (range, 1.4 to 36.1). The median time to response was 76 days (range, 20 to 137). Six of the 11 responding patients had functioning tumors and three had received prior therapy (chemotherapy, one; o,p’DDD, two). Sites of objective disease response included the following: lung, five patients; retroperitoneal mass, three patients; liver and lung, two patients; liver, one patient. In the six patients with functioning tumors, both objective and biochemical responses were noted. In the five ineligible patients, one partial tumor regression was observed. The following fac- tors were examined for their relation to response: risk sta- tus, functional status of the tumor, prior therapy, surgical status, and performance status. None of these were as- sociated with response.
The median survival of the 37 eligible patients was 11.8 months (range 0.2 to 51.2), and for the five ineligible pa- tients was 15.0 months (range, 0.1 to 20.0). The following factors were examined for their relation to survival: risk status, functional status of the tumor, prior therapy, sur- gical status, performance status, and asynchronous met- astatic disease (Table 2). A significant survival advantage was found for patients with the following: (1) prior surgical removal of the primary tumor or bulky disease (P = .0005); (2) performance status of 0 to 1 (P = . 04); and (3)
| No. of Patients | Median Survival (months) | P | |
|---|---|---|---|
| CDDP dose | |||
| 100 mg/m2 | 29 | 11.6 | |
| 75 mg/m2 | 8 | 31.5 | .17 |
| Prior therapy | |||
| Yes | 12 | 21.1 | |
| No | 25 | 10.9 | .11 |
| Functional status | |||
| Yes | 15 | 10.9 | |
| No | 18 | 19.4 | .34 |
| Prior adrenal/debulking surgery | |||
| Yes | 28 | 20.5 | |
| No | 9 | 7.0 | .0005 |
| Performance status | |||
| 0-1 | 27 | 15.6 | |
| ≥ 2 | 10 | 7.0 | .04 |
| Type of metastatic disease | |||
| Synchronous | 20 | 10.9 | |
| Asynchronous | 17 | 15.6 | .04 |
| Time of chemotherapy initiation | |||
| < 3 months from metastatic disease | 23 | 10.5 | |
| ≥ 3 months from metastatic disease | 14 | 21.9 | .01 |
presence of asynchronous metastatic disease (P = . 04). These groups were related. A performance status of 0 to 1 was associated with both asynchronous metastatic dis- ease (P = . 054) and prior surgery (P = . 027). A Cox model using stepwise selection criteria indicated that the only independent factors that were predictors for improved survival were prior surgery and a performance status of 0 to 1.
The toxicity of the CDDP and o,p’DDD combination was moderate to severe. The median number of cycles administered was three, with a range of one to nine. Thirty- six patients were assessable for toxicity, and 16 (47%) dis- continued therapy because of side effects. The toxicity experienced by this patient group is outlined in Table 3. Seventeen (47%) patients experienced ≥ grade 3 toxicity. The most common side effects were gastrointestinal (97%), renal (32%), and neurologic (27%). The gastrointestinal side effects were nausea and vomiting in most patients and were diarrhea in 11%. Renal toxicity included ele- vation of creatinine and/or BUN levels and was generally moderate and resolved; however, a single patient required hemodialysis. The neurologic toxicity consisted of par- esthesias, weakness, and, in one patient, somnolence. He- matologic toxicity was observed: 42% of patients devel- oped leukopenia, and one developed life-threatening thrombocytopenia. No treatment-related deaths were re- ported.
DISCUSSION
This study represents the first SWOG evaluation of chemotherapy in patients with adrenocortical carcinoma. The rarity of this tumor (incidence two per 1 million peo- ple per year) requires a cooperative approach to assess new therapeutic modalities in a reasonable time period. The present study investigated the utility of a combined approach employing chemotherapy with CDDP and o,p’DDD, a compound capable of producing selective ad- renocortical necrosis. A response rate of 30% was noted, and therefore, this combination seems to have modest antitumor activity. Importantly, clinical responses oc- curred regardless of prior therapy, previous surgery, or functional status.
These observations confirm previous anecdotal re- ports12,13 in which responses to CDDP were observed. It is unclear whether the addition of o,p’DDD increased the effectiveness of CDDP, but certainly the combination seems to have had additive toxicity. Substantial neurologic and gastrointestinal toxicity occurred and led to premature discontinuation of therapy in 47% of patients. The severity of the nausea and vomiting associated with this combi- nation potentially could be decreased by use of an anti- emetic, such as ondansetron hydrochloride.
The majority of responses were partial, with only one complete regression, but in many of the responding pa- tients substantial symptomatic improvement was re- ported. Recent reports that used other CDDP combina- tions, such as etoposide (VP-16) and CDDP15 with or without bleomycin,16 also have noted responses. Addi- tionally, the combination of CDDP, doxorubicin, and cy- clophosphamide has been reported to produce responses in this patient group.17 Whether such combinations will
| Type of Toxicity | Toxicity Grade (%) | |
|---|---|---|
| 1-2 | 3-4 | |
| Gastrointestinal | ||
| Nausea and vomiting | 75t | 22 |
| Diarrhea | 11 | - |
| Mucositis | 6 | - |
| Bilirubin elevation | 6 | - |
| Hematologic | ||
| Anemia | 8 | 8 |
| Leukopenia | 36 | 6 |
| Thrombocytopenia | - | 3 |
| Renal (elevated creatinine) | 17 | 8 |
| Neurologic | ||
| Peripheral neuropathy | 3 | 6 |
| Myalgias, weakness | 17 | 6 |
*Maximal toxicity experienced during treatment. tPercentage of eligible (n = 37) patients.
result in improved response rates requires further study. An ongoing SWOG trial that uses CDDP, VP-16, and o,p’DDD in patients with metastatic adrenal cortical car- cinoma is addressing this issue.
The median survival observed for patients in this study was 11.8 months, and is similar to that reported previ- ously,4,18 which ranged from 7.0 to 24.0 months. Kara- kousis et al19 reported that resection of the primary tumor in patients with metastatic disease was associated with improved survival. In the present trial, however, a history of prior resection of the primary tumor was also associated with better performance status. Also, there was no cor- relation between functional status of the tumor, initial CDDP dose (risk status), and patient survival. The median survival periods of responding and nonresponding patients (26.3 and 10.9 months, respectively) were not significantly different (P = . 11). These data suggest a natural history for metastatic adrenal cancer that is similar to that of other endocrine tumors, with prolonged survival occurring in occasional patients. The application of cytotoxic treat- ment can therefore be reserved for patients with symp- tomatic and/or progressive disease and poor prognostic factors.
As in other trials, pathologic review to determine patient eligibility may be important. Because many patients have adrenal masses found on computed tomography and metastatic lesions, pathologic confirmation of adrenal cancer is necessary. No pathologic features associated with improved response or survival were found in the present study, but only 27 patients had material from the primary tumors submitted. Most adrenal cortical carcinomas can be readily recognized in a surgically resected adrenal gland, but in some instances there can be difficulty in distin- guishing carcinoma from an adenoma. The gross and his- tologic features used to identify a malignant adrenal cor- tical tumor include invasion of contiguous structures, capsular and vascular invasion, high mitotic rate, atypical mitoses, gross and/or microscopic necrosis, and the pres- ence of a diffuse growth pattern. These features, however, are not pathognomonic of carcinoma, as is the develop- ment of metastases. In this study, evidence of metastatic or recurrent adrenal carcinoma was obtained more fre- quently on a clinical than on a pathologic basis.
The differential diagnosis of adrenal cortical carcinoma includes hepatocellular carcinoma, renal cell carcinoma, metastatic lung carcinoma, and malignant melanoma. Indeed, in a limited specimen, such as a biopsy sample or a fine-needle aspiration, it may not be possible to de- termine if a malignant tumor is primary or metastatic to the adrenal gland. In our trial, 10 of the eligible patients
had a diagnosis made on the basis of a biopsy of metastatic lesions.
In conclusion, the present study indicates that CDDP is an agent active in adrenal carcinoma. The study design, however, does not permit separation of the effects of
CDDP from those of o,p’DDD. The use of chemotherapy in patients with progressive or symptomatic disease and/ or those with synchronous metastases is reasonable. Ad- ditional trials are required to assess the utility of other agents, such as VP-16.
REFERENCES
1. National Cancer Institute: Third National Cancer survey: In- cidence data. Natl Cancer Inst Monogr 41, 1975, pp 388-393
2. Hutter AM, Kayhoe DE: Adrenal cortical carcinoma-Clinical features of 138 patients. Am J Med 41:572-580, 1966
3. MacFarland DA: Cancer of the adrenal cortex: The natural history, prognosis and treatment in a study of 55 cases. Ann R Coll Surg Engl 23:155-164, 1958
4. Laposett MG, Hertz R, Roso GT: Clinical and pathophysiologie aspects of adrenocortical carcinoma. Am J Med 35:374-383, 1963
5. Sommers SC: Adrenal glands, in Anderson WAD (ed): Pa- thology, vol 2 (ed 6). St Louis, MO, Mosby, 1970, pp 1464-1487
6. Lubitz JR, Freeman L, Okun R: Treatment of inoperable ad- renal cortical carcinoma with mitotane (o,p’DDD). Int Pharm J 17: 86-93, 1974
7. Hag MM, Legha SS, Samaan NA, et al: Cytotoxic chemotherapy in adrenal cortical carcinoma. Can Treat Rep 64:909-913, 1980
8. van Slooten H, Van Oosterom AT: CAP (Cyclophosphamide, doxorubicin and cisplatin) regimen in adrenal cortical carcinoma. Can Treat Rep 67:377-379, 1983
9. DeVita VT, Carbonne PP, Owens AH, et al: Clinical trials with 1,3-bis (2-chloroethyl)-1-nitrosourea, NSC-409962. Cancer Res 25: 1876-1881, 1965
10. Ostuni JA, Roginsky MS: Metastatic adrenal cortical carci- noma-Documented cure with combined chemotherapy. Arch Intern Med 135:1257-1258, 1975
11. Friedman MA, Ignoffo RJ, Resser KJ, et al: Combination 5FU + moderate dose methotrexate (MTX) + leucovorin (L)-Phase I- Il study of patients with disseminated colorectal cancer. Proc Am Assoc Cancer Res/Am Soc Clin Oncol 20:24, 1979 (abstr)
12. Grapski RT, Lopez JA, Krikorian JG, et al: Cisplatin che- motherapy of adrenocortical carcinoma. Proc Am Soc Clin Oncol 2:232, 1983 (abstr)
13. Tattersal HHN, Lander H, Bain B: Cis-platinum treatment of metastatic adrenal carcinoma. Med J Aust 1:419-421, 1980
14. Bukowski RM, Johnson KG, Peterson RF, et al: Phase II trial of combination chemotherapy in patients with metastatic carcinoid tumors: A SWOG study. Cancer 60:2891-2896, 1987
15. Johnson DH, Greco FA: Treatment of metastatic adrenal cor- tical carcinoma with cisplatin and etoposide (VP-16). Cancer 58: 2198-2202, 1986
16. Hesketh PJ, McCaffery RP, Finkel HE, et al: Cisplatin-based treatment of adrenocortical carcinoma. Cancer Treat Rep 71:222- 224, 1987
17. Schlumberger M, Ostronoff M, Bellaiche M, et al: 5-Fluoro- uracil, doxorubicin, and cisplatin regimen in adrenal cortical carci- noma. Cancer 61:1492-1494, 1988
18. Nador S, Hickey RC, Sellin RV, et al: Adrenal cortical car- cinoma: A study of 77 cases. Cancer 52:707-711, 1983
19. Karakousis CP, Uribe J, Moore R: Adrenal adenocarcinomas: Diagnosis and management. J Surg Oncol 16:385-389, 1981