Phase I Study of Toremifene in Patients With Advanced Cancer By John T. Hamm, Douglas C. Tormey, Peter C. Kohler, Daniel Hailler, Mark Green, and Irving Shemano A phase I multicenter evaluation of a novel antiestrogen, toremifene, was undertaken in postmenopausal women with various advanced difficult-to-treat malignancies. One hundred and seven women were treated at one of six dosage levels (10, 20, 40, 60, 200, or 400 mg/d orally) for at least 8 weeks. Weekly evaluations for toxicity were conducted. The most common side effects were nausea (31%), vomiting (12%), and hot flashes (29%). Five patients were removed from the study for possible adverse reactions: three patients experienced hypercalcemia; one experienced tremu-
HORMONAL
THERAPY has been a main-
stay of breast cancer treatment for a century. Approximately 30% of patients with advanced breast cancer respond to hormonal manipulation.' Of patients with estrogen receptor (ER)- and/or progesterone receptor (PR)-positive tumors, approximately 60% responded to treatment.2 In 1974, the antiestrogen tamoxifen was introduced. Antiestrogens bind to ERs in tumor cells, impairing the growth of estrogen-dependent tumors.3 Tamoxifen decreases the secretion of the autocrine growth factor transforming growth factor-alpha and increases the secretion of the growth inhibitor transforming growth factor-beta. Tamoxifen is widely used to treat breast cancer patients with advanced disease and in the adjuvant setting3 in postmenopausal patients. However, not all patients respond to tamoxifen, and most individuals who initially respond eventually become resistant to treatment. 4
From the Department of Medicine, University of Louisville, Louisville, KY; Departmentof Medicine, University of Wisconsm, Madison, WI; Department of Medicine, University of Pennsylvania, Philadelphia, PA; Department of Medicine, Universityof Cahfornia,San Diego, CA; and Adna Laboratones, Columbus, OH. Submitted December 27, 1990; accepted May 16, 1991. Study sponsored by Adria Laboratones,Columbus, OH. Address repnnt requests to John T. Hamm, MD, Division of MedicalOncology/Hematology, 529 South Jackson St, University of Louisville, Louisville, KY 40292. o 1991 byAmerican Society of Clinical Oncology. 0732-183X/91/0911-0001$3.00/0
2036
lousness, fatigue, and inability to think clearly; and one had vaginal bleeding. Twelve patients died while on study, 11 with disease progression and one with a pulmonary embolus. Sex hormone-binding globulin (SHBG) levels increased and there was a modest decline in serum antithrombin III levels. Four of 48 assessable patients had partial responses: three with breast cancer and onewith endometrial cancer. Toremifene was generally well tolerated at the doses tested. J Clin Oncol 9:2036-2041. o 1991 by American Society of Clinical Oncology. There also is concern over the long-term safety of tamoxifen because it is a partial estrogen agonist. In particular, there is a possible increased risk of uterine carcinoma and hepatocellular carcinoma, which may be related to long-term estrogen exposure. Hepatic hyperplasia and hepatocellular carcinomas have been noted in rats treated for 6 months with very high doses (5 mg/kg/d) of tamoxifen.5 One large clinical trial reported an increased incidence of uterine carcinomas with adjuvant tamoxifen.6 This is of even greater concern when considering long-term adjuvant therapy with an antiestrogen or in possible preventive trials. Toremifene is a triphenylethylene derivative structurally related to tamoxifen (Fig 1). The compound was synthesized and selected as part of a drug discovery program for its ability to bind to ERs in breast tissue and its activity against the MCF-7 breast cancer cell line. Toremifene inhibits the induction of rat mammary carcinoma induced by dimethylbenzanthracene (DMBA) and causes the regression of DMBA-induced tumors at a dose of 1 mg/kg/d. At equivalent antiestrogenic doses, it appears to have less estrogenic effect than tamoxifen. In 20- to 24-day-old rats and mature rats, toremifene is less uterotrophic than tamoxifen.7 High-dose toremifene (100 and 200 mg/ kg/d) has cytolytic effects in mouse ER-negative, uterine sarcoma tumors. This is not observed with high-dose tamoxifen.' In animals treated with toremifene 48 mg/kg/d for 26 weeks, hyperplastic nodules of the liver were not observed. High-dose
Journal of Clinical Oncology, Vol 9, No 11 (November), 1991: pp 2036-2041
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2037
PHASE I STUDY OF TOREMIFENE METHODS PatientSelection
Fig 1. Chemical structures of toremifene, its two predominant metabolites and tamoxifen. Toremifene: R', OCH,CH 2 N(CH 3 )2 ; R", H; R"', CH 2 CL. N-desmethyltoremifene: R', OCH 2 CHINHCH,; R", H; RI", CH 2CL. Tamoxifen: R', OCHCH2N(CH,),; R", H; R"', H. 4-Hydroxyloremifene: R', OCH2 CH 2N(CH 3 )2 ; R", OH; R"', CH 2 CL.
The phase I trial was an open-label, randomized trial in postmenopausal women with various tumor types. The population was limited to females, the target for this drug, and to non-child-bearing individuals (ie, at least 1 year postmenopausal or prior oophorectomy) because teratogenicity studies have not been completed. Patients must have had advanced tumor for which no standard therapy has proven to be effective or must have been previously treated, with therapy failure. Patients eligible for the phase I trial who had previously been treated with tamoxifen must have been off therapy for at least 3 months. Patients previously treated with cytotoxic agents or other hormone therapy must have been off therapy for at least 3 weeks or until recovery from the toxic effects of therapy had occurred. Patients must have had a WBC count > 3,500/!iL, a neutrophil count > 1,500/!1L, a platelet count > 100,000/ptL, and normal liver and renal function. Performance status must have been Eastern Cooperative Oncology Group grade 0, 1, or 2. The patient must have been able to comply with scheduled study visits and follow-up tests. Written consent was obtained from each patient for participation in the trial.
Design
tamoxifen has been observed to cause hyperplastic liver nodules in the same model.8 Pharmacokinetic data in female volunteers demonstrated good absorption after oral administration of doses ranging from 3 to 680 mg.9 The peak serum concentration was reached in 4 hours, and the elimination half-life-3 was 5 days. At a dose of 60 mg/d, steady state was reached in 6 weeks, with a mean steady-state plasma concentration of 0.65 mcg/mL, with elimination primarily in the feces (70%). Two active metabolites are formed: N-desmethyl and 4-hydroxy derivatives. Toremifene is strongly bound to protein. There was a modest decrease in serum antithrombin III levels, a decline in serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH), and a rise in sex hormone-binding globulin (SHBG). The synthesis of toremifene originated with the concept that an analogue of tamoxifen with broadened clinical utility could be developed. Based on the differences noted in animal models discussed above, the hypothesis was that toremifene is qualitatively different, in part, from tamoxifen and may be more broadly useful. Thus, a phase I study was initiated to determine tolerance to toremifene over a range of doses active in animal models, assess endocrine effects, and determine steady-
state pharmacokinetics.
Of the 108 patients enrolled in this study, 107 were treated. The original intention was to accrue 60 assessable patients, but many patients declined the testing necessary for the steady-state pharmacokinetics portion of the study. Therefore, the trial was extended to achieve an enrollment of four to six patients per dose level who were assessable for steady-state pharmacokinetics. When this trial was designed, preliminary toxicity data on toremifene was available from European trials.' It was considered to be relatively nontoxic; therefore, in this trial the dosage was not escalated sequentially. Preliminary data from these trials suggested response rates of 54% and 57% in postmenopausal ER-positive patients treated with toremifene 60 mg/d. Therefore, doses were designed to test a range from one sixth to greater than sixfold the anticipated clinically useful dose. Males were excluded because the intended treatment population is individuals with breast cancer. Patients were randomly assigned to one of six dosage groups. The dosage range selected included doses that exerted an antiestrogenic effect in preliminary animal and human studies.7' To facilitate the collection of data for steady-state pharmacokinetics, data collection was restricted to three study sites. Patients were randomized to dosage levels of 20, 60, and 400 mg/d. The other sites randomized dosages patients to the 10, 40, and 200 mg/d dose levels. Patient dosages were randomized by the Biostatistics Department at Adria Laboratories according to a computer-generated code. The pharmacokinetic portion of the study is reported separately."' Pretreatment evaluation included a medical history, physical examination, ophthalmologic examination, performance status, chest x-ray, laboratory profiles (hematology, serum chemistry, LH, FSH, estradiol, estrone, thyroxine [T4], cortisol, prolactin, and antithrombin III), ECG, and
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2038
HAMM ET AL
tumor measurements. The ophthalmic exam consisted of ophthalmoscopy, slit lamp exam, and visual acuity tests by a qualified ophthalmologist. While on study, assessments included complete blood cell count once weekly for 4 weeks, at 8 weeks, and every 12 weeks thereafter. Ocular examinations and ECGs were repeated after 8 weeks. The following tests were done at 4 and 8 weeks and every 12 weeks thereafter: performance status, chest x-ray, laboratory profiles, toxicity assessments, and tumor assessment in patients with assessable disease. Blood was sampled weekly for timed steady-state, and multiple blood samples were collected after the final dose for terminal pharmacokinetics. Side effects were determined once weekly by asking patients if they had experienced any ill effects since the last visit. These were reported as mild (grade I), moderate (grade II), or severe (grade III). To assess the dosage of toremifene, which inhibits estrogen effect on vaginal epithelium, a vaginal cornification assay was performed. At baseline and after six weeks of toremifene treatment, two vaginal smears were obtained. A transdermal estradiol patch (Estraderm; Ciba-Geigy, Summit, NJ) that administered 100 ýtg/d was applied to the upper arm. This was changed at 3.5, 7, and 10 days. After 14 days, vaginal smears were taken again. The smears were reviewed by a reference cytologist and scored for maturation index.
Table 1. Patientt Characteristics Mean age, years (range) Race White Black Other Performance status 0 1 2 No data Primary Breast Ovary Uterine Unknown Lung Gastrointestinal Nasopharynx Melanoma Sarcoma Previous therapy Surgery Chemotherapy Radiation Hormonal Immunotherapy
58 (25-80) 91 8 8 54 34 10 9 74 12 5 6 3 3 2 1 1 101 88 69 63 8
Treatment Patients were assigned to one of six dosage groups (10, 20, 40, 60, 200, or 400 mg/d) by the method described above. Toremifene was administered orally once daily for at least 8 weeks in the absence of disease progression or untoward toxicity. If the disease remained stable or responded to therapy, toremifene was continued indefinitely until progression or untoward toxicity intervened.
RESULTS
One hundred seven patients were randomized and treated in the study. Patient characteristics are summarized in Table 1. The mean age was 58 years, with a range of 25 to 80 years. Sixty-eight percent of the patients had advanced breast can-
cer; the remainder had bone, lung, endometrial,
bile duct, ovarian, colon, nasal, nasopharyngeal, melanoma, or duodenal cancer and unknown primary malignancies. Forty-five percent of the patients had measurable disease.
Toxicity Five patients were removed from the study because of possible toremifene-related adverse reactions (Table 2). These reactions occurred at various doses and did not appear to be doserelated. The reactions included hypercalcemia,
bone pain, tremulousness, fatigue, urticaria, and vaginal bleeding. While on study 12 patients died: 11 of progressive disease and one of an acute pulmonary embolus (Table 3). Toremifene was generally well tolerated at all doses, as summarized in Table 4. The most common adverse effects were nausea and hot flashes. Nausea occurred in 31% of patients, vomiting in 12%, with severe nausea or vomiting in two patients. Patient complaints of abdominal pain or discomfort (10%), anorexia (9%), diarrhea (4%), constipation (3%), and increased appetite (2%) were less common. Hot flashes were noted in 29% of patients; this occurred at all dose levels and did not appear to be dose-related. Vaginal discharge Table 2. Patients Removed From Study for Adverse Reactions Toremifene Dose (mg/d) 20 60 200 400 400
Adverse Reaction Hypercalcemia Bone pain, hypercalcemia Tremulousness, fatigue, inability to think clearly Urticaria, hypercalcemia, bone pain Heartburn, vaginal bleeding
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
Days on Study 12 69 19 26 118
2039
PHASE I STUDY OF TOREMIFENE
There was no consistent change in cortisol, prolactin, estrone, estradiol, T3, T4, or thyroid-stimulating hormone levels. There was a modest decline in the serum antithrombin III level (Table 6). No consistent change in other serum chemistries or hematologic values was noted. Thirty-five patients completed the evaluation of estrogen blockage on the vaginal epithelium (Table 7). Only one of four patients assessed at 10 mg/d demonstrated estrogen blockage, compared with 22 of 31 patients treated with > 20 mg/d. Notably, three of six patients treated at the highest dose of 400 mg/d demonstrated estrogen blockade.
Table 3. Patient Deaths on Study Toremifene Dose (mg/d)
10 20 40 60 200
400
Days on Study
31 12 28 165 15 22 163 162 20 33 32 150
Most Probable Cause
Pneumonia and progressive disease Disease progression Disease progression Disease progression Disease progression Disease progression Brain metastases Disease progression Acute pulmonary embolus considered unrelated to drug Disease progression Disease progression Disease progression
Response Four of 48 assessable patients demonstrated partial responses. Three of these responding patients had metastatic breast cancer and received toremifene 200 mg/d. Disease in two of these patients had previously responded and then progressed while on tamoxifen; in the third patient, disease had progressed on chemotherapy with cyclophosphamide, methotrexate, and fluorouracil. The fourth response occurred in a patient with advanced endometrial carcinoma treated with toremifene 400 mg/d.
(8.4%) and vaginal bleeding (2.8%) also occurred. CNS symptoms reported included dizziness/ vertigo (10%), lethargy/fatigue (10%), and headaches (9%). Weight gain was not noted as an adverse reaction. Three patients with breast cancer and bone metastasis developed bone pain and hypercalcemia and were taken off study. Ophthamologic toxicities were closely monitored. Dry eye (reduced tearing) was reported in three patients (at dose levels of 60, 200, and 400 mg/d). Two patients (at 60 mg/d) had mild cataracts noted at the 8-week exam that were not reported in the entry exam. The possibility of ophthamalogic toxicities is being carefully monitored in further studies. Although serum hormone levels varied considerably during the eight weeks, a mild decline in LH and FSH was noted at higher doses (Table 5). SHBG levels increased during therapy (Table 5).
DISCUSSION
Toremifene was generally well tolerated over the wide range of doses tested. The most common side effects were nausea, vomiting, and hot flashes. Although the frequency of nausea appeared to be higher than is commonly reported with tamoxifen,
Table 4. Adverse Effects Dose (mg/d)
Nausea Vomiting Abdominal discomfort Anorexia Hot flashes Vaginal discharge Vaginal bleeding Dizziness/vertigo Lethargy/fatigue Headaches
10 (n = 11)
20 (n = 26)
40 in = 14)
60 (n = 22)
200 (n = 10)
400 (n = 24)
Total (N = 107)
3 0 2 0 3 0 0 0 0 0
8 4 2 2 6 0 1 0 4 3
4 2 2 1 5 2 1 2 0 2
4 2 0 5 6 3 0 3 3 1
5 2 0 0 6 2 0 1 3 2
9 3 2 1 5 1 1 5 1 2
33 13 8 9 31 8 3 11 11 10
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2040
HAMM ET AL Table 5. Peak Change in Gonadotropins After Therapy
Dose (mg/d) 10 20 40 60 200 400
Baseline 61 76 56 58 74 48
+- 38 t 26 ± 28 - 30 ± 43 + 33
3.8 + -8.7 + 13 + -17 -39 ± -16 ±
SHBG (nmol/L)
LH(mlU/mL)
FSH (mlU/mL) Change
Baseline
12 11 17 5 35 27
21 22 t 15 ± 26+ 20 ± 19 _
12 8.5 5 14 14 12
Change 5.3 1.9 3.3 -5.3 -10.1 -7 8
± t + ± ± ±
7.4 13 3.8 5.2 12.4 9.3
Baseline 62 52 32 40 57 55
Change
t 25 + 31 13 -14 ± 16 t 20
18 t 24 + 21 40 t 58 t 69 +
17 30 8 31 12 27
NOTE. Values are means ± SD. FSH, LH, and SHBG were measured pretreatment and weekly for 8 weeks. The results were expressed as the peak change of each during treatment, compared with baseline values.
this may be due to the trial population, which consisted of patients with advanced, progressive disease. A current phase III trial comparing torem-
ifene to tamoxifen may determine if nausea is more frequent with toremifene. A tumor-flare reaction with increasing bone pain and hypercalcemia was noted in three patients. These patients were taken off study; thus, it is not possible to determine if this predicted a tumor response. As reported with tamoxifen, 111' 2 a decrease in serum antithrombin III and an increase in SHBG were noted in this study. As this study was restricted to postmenopausal females, the toxicity profile in premenopausal females and males is unknown. The estrogen blockade test suggests estrogeninduced vaginal epithelial maturation is not blocked with toremifene at 10 mg/d but is at doses > 20 mg/d. The finding that only three of six patients in the 400 mg/d group were estrogen-blocked may be random variation or may suggest that at higher doses estrogen effects begin to dominate. Two patients (at 60 mg/d) were noted at week 8 to have early cataracts that were not noted in the entry examination. These two cases were reviewed by the study ophthamalogist and were believed to have been overlooked on the initial exam. No data have related tamoxifen to cataracts. This is being
examined in follow-up studies with more patients
and longer-term use. As this was a phase I study of tolerance, disease evaluation was not an entry requirement, and only 48 patients were considered assessable. Only four partial responses were seen in this group. This was a heavily pretreated population, not selected by criteria that would indicate responsiveness to hormonal therapy. The partial responses in two patients who previously responded to tamoxifen are interesting, as one would expect antiestrogens to
exert an antitumor effect by the same mechanism. The three patients with breast cancer and tumor responses were treated with toremifene 200 mg/d. From these limited numbers of patients and responses, it is not possible to determine if this represents a dose-response effect, but in studies with tamoxifen, no definitive dose-response relationship has been established for higher doses. The responsiveness of tumors in women with breast cancer progressing on tamoxifen is undergoing evaluation in phase II trials."3 A phase III trial
comparing tamoxifen with toremifene in postmenopausal patients with breast cancer is also underway. Table 7. Vaginal Cornification Assay of Estrogen Blockade
Table 6. Antithrombin III (Functional Assay) Dose
No.
(mg/d)
Patients
Baseline*
Peak Change
10 20 40 60 200 400
5 15 7 16 4 16
119 12 114 17 100 _+16 113 13 117 9 116 22
-5 -15 -4.6 -15 -12 -12
NOTE. Values are means : SD. *Normal levels, 85%to 120%.
± ± ± + +
14 10 21 11 3 18
Toremifene Dose (mg/d)
Estrogen Blockade*
10 20 40 60 200 400
1/4 7/8 3/8 6/6 3/3 3/6
*Superficial cell count did not increase by more than 10 cells per 200 cell field from week 6 to week 8. Values indicate number of patients with estrogen blockade per number of patients tested in group.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2041
PHASE I STUDY OF TOREMIFENE APPENDIX
The following were additionalparticipatinginvestigators: C. Benz, University of California, San Francisco; C.K. Osborne, University of Texas, San Antonio; K. Pandya, University of Rochester, NY; C. Vogel, Miami, FL; and D. Hayes, Dana-Farber Cancer Institute, Boston, MA.
REFERENCES 1. Furr BJA, Jordan VD: The pharmacology and clinical uses of tamoxifen. Pharmacol Ther 25:127-205, 1984 2. Legha SS, Davis HL, Muggia FM: Hormonal therapy of breast cancer: New approaches and concepts. Ann Intern Med 88:69-77, 1978 3. Lerner LJ, Jordan VC: Development of antiestrogens and their use in breast cancer: Eighth Cain Memorial Award lecture. Cancer Res 50:4177-4189, 1990 4. Hamm JT, Allegra JC: Factors in hormonal responsiveness: Loss of hormonal responsiveness in cancer in Stoll BA (ed): Endocrine Management of Cancer, Biological Bases. London, UK, Karger, 1988, pp 61-71 5. Rattel B, Loser R, Dahma EG, et al: Comparative toxicology of droloxifene (3-OH tamoxifen) and tamoxifen: Hepatocellular carcinomas induced by tamoxifen. Proceedings of the International Association of Cancer Research Meeting, Miami, FL, 1987 6. Fornander T, Rutquist LE, Cedermark B, et al: Adjuvant tamoxifen in early breast cancer: Occurrence of new primary cancers. Lancet 1:117-120, 1989 7. Kallio S, Kangas L, Blanco G, et al: A new triphenyleth-
ylene compound, Fc-1157a I. Hormonal effects. Cancer Chemother Pharmacol 17:103-108, 1986 8. Kangas L, Nieminen AL, Blanco G, et al: A new triphenylethylene compound, Fc-1157a II. Antitumor effects. Cancer Chemother Pharmacol 17:109-113, 1986 9. Toremifene (Fcl157a) clinical brochure. Columbus, OH, Adria Laboratories, 1989 10. Wiebe VJ, Benz CC, Shemano I, et al: Pharmacokinetics of toremifene and its metabolites in patients with advanced breast cancer. Cancer Chemother Pharmacol 25:247-251, 1990 11. Legha SS, Powell K, Buzdar AU, et al: Tamoxifeninduced hypercalcemia in breast cancer. Cancer 47:28032806, 1981 12. Enck RE, Rios CN: Tamoxifen treatment of metastatic breast cancer and antithrombin III levels. Cancer 53:2607-2609, 1984 13. Valvarra R, Pyrhonen S, Heikkinen M, et al: Toremifene, a new antiestrogenic compound for treatment of advanced breast cancer. Phase II study. Eur J Cancer Clin Oncol 24:785-790, 1988
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
HORMONAL
THERAPY has been a main-
stay of breast cancer treatment for a century. Approximately 30% of patients with advanced breast cancer respond to hormonal manipulation.' Of patients with estrogen receptor (ER)- and/or progesterone receptor (PR)-positive tumors, approximately 60% responded to treatment.2 In 1974, the antiestrogen tamoxifen was introduced. Antiestrogens bind to ERs in tumor cells, impairing the growth of estrogen-dependent tumors.3 Tamoxifen decreases the secretion of the autocrine growth factor transforming growth factor-alpha and increases the secretion of the growth inhibitor transforming growth factor-beta. Tamoxifen is widely used to treat breast cancer patients with advanced disease and in the adjuvant setting3 in postmenopausal patients. However, not all patients respond to tamoxifen, and most individuals who initially respond eventually become resistant to treatment. 4
From the Department of Medicine, University of Louisville, Louisville, KY; Departmentof Medicine, University of Wisconsm, Madison, WI; Department of Medicine, University of Pennsylvania, Philadelphia, PA; Department of Medicine, Universityof Cahfornia,San Diego, CA; and Adna Laboratones, Columbus, OH. Submitted December 27, 1990; accepted May 16, 1991. Study sponsored by Adria Laboratones,Columbus, OH. Address repnnt requests to John T. Hamm, MD, Division of MedicalOncology/Hematology, 529 South Jackson St, University of Louisville, Louisville, KY 40292. o 1991 byAmerican Society of Clinical Oncology. 0732-183X/91/0911-0001$3.00/0
2036
lousness, fatigue, and inability to think clearly; and one had vaginal bleeding. Twelve patients died while on study, 11 with disease progression and one with a pulmonary embolus. Sex hormone-binding globulin (SHBG) levels increased and there was a modest decline in serum antithrombin III levels. Four of 48 assessable patients had partial responses: three with breast cancer and onewith endometrial cancer. Toremifene was generally well tolerated at the doses tested. J Clin Oncol 9:2036-2041. o 1991 by American Society of Clinical Oncology. There also is concern over the long-term safety of tamoxifen because it is a partial estrogen agonist. In particular, there is a possible increased risk of uterine carcinoma and hepatocellular carcinoma, which may be related to long-term estrogen exposure. Hepatic hyperplasia and hepatocellular carcinomas have been noted in rats treated for 6 months with very high doses (5 mg/kg/d) of tamoxifen.5 One large clinical trial reported an increased incidence of uterine carcinomas with adjuvant tamoxifen.6 This is of even greater concern when considering long-term adjuvant therapy with an antiestrogen or in possible preventive trials. Toremifene is a triphenylethylene derivative structurally related to tamoxifen (Fig 1). The compound was synthesized and selected as part of a drug discovery program for its ability to bind to ERs in breast tissue and its activity against the MCF-7 breast cancer cell line. Toremifene inhibits the induction of rat mammary carcinoma induced by dimethylbenzanthracene (DMBA) and causes the regression of DMBA-induced tumors at a dose of 1 mg/kg/d. At equivalent antiestrogenic doses, it appears to have less estrogenic effect than tamoxifen. In 20- to 24-day-old rats and mature rats, toremifene is less uterotrophic than tamoxifen.7 High-dose toremifene (100 and 200 mg/ kg/d) has cytolytic effects in mouse ER-negative, uterine sarcoma tumors. This is not observed with high-dose tamoxifen.' In animals treated with toremifene 48 mg/kg/d for 26 weeks, hyperplastic nodules of the liver were not observed. High-dose
Journal of Clinical Oncology, Vol 9, No 11 (November), 1991: pp 2036-2041
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2037
PHASE I STUDY OF TOREMIFENE METHODS PatientSelection
Fig 1. Chemical structures of toremifene, its two predominant metabolites and tamoxifen. Toremifene: R', OCH,CH 2 N(CH 3 )2 ; R", H; R"', CH 2 CL. N-desmethyltoremifene: R', OCH 2 CHINHCH,; R", H; RI", CH 2CL. Tamoxifen: R', OCHCH2N(CH,),; R", H; R"', H. 4-Hydroxyloremifene: R', OCH2 CH 2N(CH 3 )2 ; R", OH; R"', CH 2 CL.
The phase I trial was an open-label, randomized trial in postmenopausal women with various tumor types. The population was limited to females, the target for this drug, and to non-child-bearing individuals (ie, at least 1 year postmenopausal or prior oophorectomy) because teratogenicity studies have not been completed. Patients must have had advanced tumor for which no standard therapy has proven to be effective or must have been previously treated, with therapy failure. Patients eligible for the phase I trial who had previously been treated with tamoxifen must have been off therapy for at least 3 months. Patients previously treated with cytotoxic agents or other hormone therapy must have been off therapy for at least 3 weeks or until recovery from the toxic effects of therapy had occurred. Patients must have had a WBC count > 3,500/!iL, a neutrophil count > 1,500/!1L, a platelet count > 100,000/ptL, and normal liver and renal function. Performance status must have been Eastern Cooperative Oncology Group grade 0, 1, or 2. The patient must have been able to comply with scheduled study visits and follow-up tests. Written consent was obtained from each patient for participation in the trial.
Design
tamoxifen has been observed to cause hyperplastic liver nodules in the same model.8 Pharmacokinetic data in female volunteers demonstrated good absorption after oral administration of doses ranging from 3 to 680 mg.9 The peak serum concentration was reached in 4 hours, and the elimination half-life-3 was 5 days. At a dose of 60 mg/d, steady state was reached in 6 weeks, with a mean steady-state plasma concentration of 0.65 mcg/mL, with elimination primarily in the feces (70%). Two active metabolites are formed: N-desmethyl and 4-hydroxy derivatives. Toremifene is strongly bound to protein. There was a modest decrease in serum antithrombin III levels, a decline in serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH), and a rise in sex hormone-binding globulin (SHBG). The synthesis of toremifene originated with the concept that an analogue of tamoxifen with broadened clinical utility could be developed. Based on the differences noted in animal models discussed above, the hypothesis was that toremifene is qualitatively different, in part, from tamoxifen and may be more broadly useful. Thus, a phase I study was initiated to determine tolerance to toremifene over a range of doses active in animal models, assess endocrine effects, and determine steady-
state pharmacokinetics.
Of the 108 patients enrolled in this study, 107 were treated. The original intention was to accrue 60 assessable patients, but many patients declined the testing necessary for the steady-state pharmacokinetics portion of the study. Therefore, the trial was extended to achieve an enrollment of four to six patients per dose level who were assessable for steady-state pharmacokinetics. When this trial was designed, preliminary toxicity data on toremifene was available from European trials.' It was considered to be relatively nontoxic; therefore, in this trial the dosage was not escalated sequentially. Preliminary data from these trials suggested response rates of 54% and 57% in postmenopausal ER-positive patients treated with toremifene 60 mg/d. Therefore, doses were designed to test a range from one sixth to greater than sixfold the anticipated clinically useful dose. Males were excluded because the intended treatment population is individuals with breast cancer. Patients were randomly assigned to one of six dosage groups. The dosage range selected included doses that exerted an antiestrogenic effect in preliminary animal and human studies.7' To facilitate the collection of data for steady-state pharmacokinetics, data collection was restricted to three study sites. Patients were randomized to dosage levels of 20, 60, and 400 mg/d. The other sites randomized dosages patients to the 10, 40, and 200 mg/d dose levels. Patient dosages were randomized by the Biostatistics Department at Adria Laboratories according to a computer-generated code. The pharmacokinetic portion of the study is reported separately."' Pretreatment evaluation included a medical history, physical examination, ophthalmologic examination, performance status, chest x-ray, laboratory profiles (hematology, serum chemistry, LH, FSH, estradiol, estrone, thyroxine [T4], cortisol, prolactin, and antithrombin III), ECG, and
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2038
HAMM ET AL
tumor measurements. The ophthalmic exam consisted of ophthalmoscopy, slit lamp exam, and visual acuity tests by a qualified ophthalmologist. While on study, assessments included complete blood cell count once weekly for 4 weeks, at 8 weeks, and every 12 weeks thereafter. Ocular examinations and ECGs were repeated after 8 weeks. The following tests were done at 4 and 8 weeks and every 12 weeks thereafter: performance status, chest x-ray, laboratory profiles, toxicity assessments, and tumor assessment in patients with assessable disease. Blood was sampled weekly for timed steady-state, and multiple blood samples were collected after the final dose for terminal pharmacokinetics. Side effects were determined once weekly by asking patients if they had experienced any ill effects since the last visit. These were reported as mild (grade I), moderate (grade II), or severe (grade III). To assess the dosage of toremifene, which inhibits estrogen effect on vaginal epithelium, a vaginal cornification assay was performed. At baseline and after six weeks of toremifene treatment, two vaginal smears were obtained. A transdermal estradiol patch (Estraderm; Ciba-Geigy, Summit, NJ) that administered 100 ýtg/d was applied to the upper arm. This was changed at 3.5, 7, and 10 days. After 14 days, vaginal smears were taken again. The smears were reviewed by a reference cytologist and scored for maturation index.
Table 1. Patientt Characteristics Mean age, years (range) Race White Black Other Performance status 0 1 2 No data Primary Breast Ovary Uterine Unknown Lung Gastrointestinal Nasopharynx Melanoma Sarcoma Previous therapy Surgery Chemotherapy Radiation Hormonal Immunotherapy
58 (25-80) 91 8 8 54 34 10 9 74 12 5 6 3 3 2 1 1 101 88 69 63 8
Treatment Patients were assigned to one of six dosage groups (10, 20, 40, 60, 200, or 400 mg/d) by the method described above. Toremifene was administered orally once daily for at least 8 weeks in the absence of disease progression or untoward toxicity. If the disease remained stable or responded to therapy, toremifene was continued indefinitely until progression or untoward toxicity intervened.
RESULTS
One hundred seven patients were randomized and treated in the study. Patient characteristics are summarized in Table 1. The mean age was 58 years, with a range of 25 to 80 years. Sixty-eight percent of the patients had advanced breast can-
cer; the remainder had bone, lung, endometrial,
bile duct, ovarian, colon, nasal, nasopharyngeal, melanoma, or duodenal cancer and unknown primary malignancies. Forty-five percent of the patients had measurable disease.
Toxicity Five patients were removed from the study because of possible toremifene-related adverse reactions (Table 2). These reactions occurred at various doses and did not appear to be doserelated. The reactions included hypercalcemia,
bone pain, tremulousness, fatigue, urticaria, and vaginal bleeding. While on study 12 patients died: 11 of progressive disease and one of an acute pulmonary embolus (Table 3). Toremifene was generally well tolerated at all doses, as summarized in Table 4. The most common adverse effects were nausea and hot flashes. Nausea occurred in 31% of patients, vomiting in 12%, with severe nausea or vomiting in two patients. Patient complaints of abdominal pain or discomfort (10%), anorexia (9%), diarrhea (4%), constipation (3%), and increased appetite (2%) were less common. Hot flashes were noted in 29% of patients; this occurred at all dose levels and did not appear to be dose-related. Vaginal discharge Table 2. Patients Removed From Study for Adverse Reactions Toremifene Dose (mg/d) 20 60 200 400 400
Adverse Reaction Hypercalcemia Bone pain, hypercalcemia Tremulousness, fatigue, inability to think clearly Urticaria, hypercalcemia, bone pain Heartburn, vaginal bleeding
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
Days on Study 12 69 19 26 118
2039
PHASE I STUDY OF TOREMIFENE
There was no consistent change in cortisol, prolactin, estrone, estradiol, T3, T4, or thyroid-stimulating hormone levels. There was a modest decline in the serum antithrombin III level (Table 6). No consistent change in other serum chemistries or hematologic values was noted. Thirty-five patients completed the evaluation of estrogen blockage on the vaginal epithelium (Table 7). Only one of four patients assessed at 10 mg/d demonstrated estrogen blockage, compared with 22 of 31 patients treated with > 20 mg/d. Notably, three of six patients treated at the highest dose of 400 mg/d demonstrated estrogen blockade.
Table 3. Patient Deaths on Study Toremifene Dose (mg/d)
10 20 40 60 200
400
Days on Study
31 12 28 165 15 22 163 162 20 33 32 150
Most Probable Cause
Pneumonia and progressive disease Disease progression Disease progression Disease progression Disease progression Disease progression Brain metastases Disease progression Acute pulmonary embolus considered unrelated to drug Disease progression Disease progression Disease progression
Response Four of 48 assessable patients demonstrated partial responses. Three of these responding patients had metastatic breast cancer and received toremifene 200 mg/d. Disease in two of these patients had previously responded and then progressed while on tamoxifen; in the third patient, disease had progressed on chemotherapy with cyclophosphamide, methotrexate, and fluorouracil. The fourth response occurred in a patient with advanced endometrial carcinoma treated with toremifene 400 mg/d.
(8.4%) and vaginal bleeding (2.8%) also occurred. CNS symptoms reported included dizziness/ vertigo (10%), lethargy/fatigue (10%), and headaches (9%). Weight gain was not noted as an adverse reaction. Three patients with breast cancer and bone metastasis developed bone pain and hypercalcemia and were taken off study. Ophthamologic toxicities were closely monitored. Dry eye (reduced tearing) was reported in three patients (at dose levels of 60, 200, and 400 mg/d). Two patients (at 60 mg/d) had mild cataracts noted at the 8-week exam that were not reported in the entry exam. The possibility of ophthamalogic toxicities is being carefully monitored in further studies. Although serum hormone levels varied considerably during the eight weeks, a mild decline in LH and FSH was noted at higher doses (Table 5). SHBG levels increased during therapy (Table 5).
DISCUSSION
Toremifene was generally well tolerated over the wide range of doses tested. The most common side effects were nausea, vomiting, and hot flashes. Although the frequency of nausea appeared to be higher than is commonly reported with tamoxifen,
Table 4. Adverse Effects Dose (mg/d)
Nausea Vomiting Abdominal discomfort Anorexia Hot flashes Vaginal discharge Vaginal bleeding Dizziness/vertigo Lethargy/fatigue Headaches
10 (n = 11)
20 (n = 26)
40 in = 14)
60 (n = 22)
200 (n = 10)
400 (n = 24)
Total (N = 107)
3 0 2 0 3 0 0 0 0 0
8 4 2 2 6 0 1 0 4 3
4 2 2 1 5 2 1 2 0 2
4 2 0 5 6 3 0 3 3 1
5 2 0 0 6 2 0 1 3 2
9 3 2 1 5 1 1 5 1 2
33 13 8 9 31 8 3 11 11 10
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2040
HAMM ET AL Table 5. Peak Change in Gonadotropins After Therapy
Dose (mg/d) 10 20 40 60 200 400
Baseline 61 76 56 58 74 48
+- 38 t 26 ± 28 - 30 ± 43 + 33
3.8 + -8.7 + 13 + -17 -39 ± -16 ±
SHBG (nmol/L)
LH(mlU/mL)
FSH (mlU/mL) Change
Baseline
12 11 17 5 35 27
21 22 t 15 ± 26+ 20 ± 19 _
12 8.5 5 14 14 12
Change 5.3 1.9 3.3 -5.3 -10.1 -7 8
± t + ± ± ±
7.4 13 3.8 5.2 12.4 9.3
Baseline 62 52 32 40 57 55
Change
t 25 + 31 13 -14 ± 16 t 20
18 t 24 + 21 40 t 58 t 69 +
17 30 8 31 12 27
NOTE. Values are means ± SD. FSH, LH, and SHBG were measured pretreatment and weekly for 8 weeks. The results were expressed as the peak change of each during treatment, compared with baseline values.
this may be due to the trial population, which consisted of patients with advanced, progressive disease. A current phase III trial comparing torem-
ifene to tamoxifen may determine if nausea is more frequent with toremifene. A tumor-flare reaction with increasing bone pain and hypercalcemia was noted in three patients. These patients were taken off study; thus, it is not possible to determine if this predicted a tumor response. As reported with tamoxifen, 111' 2 a decrease in serum antithrombin III and an increase in SHBG were noted in this study. As this study was restricted to postmenopausal females, the toxicity profile in premenopausal females and males is unknown. The estrogen blockade test suggests estrogeninduced vaginal epithelial maturation is not blocked with toremifene at 10 mg/d but is at doses > 20 mg/d. The finding that only three of six patients in the 400 mg/d group were estrogen-blocked may be random variation or may suggest that at higher doses estrogen effects begin to dominate. Two patients (at 60 mg/d) were noted at week 8 to have early cataracts that were not noted in the entry examination. These two cases were reviewed by the study ophthamalogist and were believed to have been overlooked on the initial exam. No data have related tamoxifen to cataracts. This is being
examined in follow-up studies with more patients
and longer-term use. As this was a phase I study of tolerance, disease evaluation was not an entry requirement, and only 48 patients were considered assessable. Only four partial responses were seen in this group. This was a heavily pretreated population, not selected by criteria that would indicate responsiveness to hormonal therapy. The partial responses in two patients who previously responded to tamoxifen are interesting, as one would expect antiestrogens to
exert an antitumor effect by the same mechanism. The three patients with breast cancer and tumor responses were treated with toremifene 200 mg/d. From these limited numbers of patients and responses, it is not possible to determine if this represents a dose-response effect, but in studies with tamoxifen, no definitive dose-response relationship has been established for higher doses. The responsiveness of tumors in women with breast cancer progressing on tamoxifen is undergoing evaluation in phase II trials."3 A phase III trial
comparing tamoxifen with toremifene in postmenopausal patients with breast cancer is also underway. Table 7. Vaginal Cornification Assay of Estrogen Blockade
Table 6. Antithrombin III (Functional Assay) Dose
No.
(mg/d)
Patients
Baseline*
Peak Change
10 20 40 60 200 400
5 15 7 16 4 16
119 12 114 17 100 _+16 113 13 117 9 116 22
-5 -15 -4.6 -15 -12 -12
NOTE. Values are means : SD. *Normal levels, 85%to 120%.
± ± ± + +
14 10 21 11 3 18
Toremifene Dose (mg/d)
Estrogen Blockade*
10 20 40 60 200 400
1/4 7/8 3/8 6/6 3/3 3/6
*Superficial cell count did not increase by more than 10 cells per 200 cell field from week 6 to week 8. Values indicate number of patients with estrogen blockade per number of patients tested in group.
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
2041
PHASE I STUDY OF TOREMIFENE APPENDIX
The following were additionalparticipatinginvestigators: C. Benz, University of California, San Francisco; C.K. Osborne, University of Texas, San Antonio; K. Pandya, University of Rochester, NY; C. Vogel, Miami, FL; and D. Hayes, Dana-Farber Cancer Institute, Boston, MA.
REFERENCES 1. Furr BJA, Jordan VD: The pharmacology and clinical uses of tamoxifen. Pharmacol Ther 25:127-205, 1984 2. Legha SS, Davis HL, Muggia FM: Hormonal therapy of breast cancer: New approaches and concepts. Ann Intern Med 88:69-77, 1978 3. Lerner LJ, Jordan VC: Development of antiestrogens and their use in breast cancer: Eighth Cain Memorial Award lecture. Cancer Res 50:4177-4189, 1990 4. Hamm JT, Allegra JC: Factors in hormonal responsiveness: Loss of hormonal responsiveness in cancer in Stoll BA (ed): Endocrine Management of Cancer, Biological Bases. London, UK, Karger, 1988, pp 61-71 5. Rattel B, Loser R, Dahma EG, et al: Comparative toxicology of droloxifene (3-OH tamoxifen) and tamoxifen: Hepatocellular carcinomas induced by tamoxifen. Proceedings of the International Association of Cancer Research Meeting, Miami, FL, 1987 6. Fornander T, Rutquist LE, Cedermark B, et al: Adjuvant tamoxifen in early breast cancer: Occurrence of new primary cancers. Lancet 1:117-120, 1989 7. Kallio S, Kangas L, Blanco G, et al: A new triphenyleth-
ylene compound, Fc-1157a I. Hormonal effects. Cancer Chemother Pharmacol 17:103-108, 1986 8. Kangas L, Nieminen AL, Blanco G, et al: A new triphenylethylene compound, Fc-1157a II. Antitumor effects. Cancer Chemother Pharmacol 17:109-113, 1986 9. Toremifene (Fcl157a) clinical brochure. Columbus, OH, Adria Laboratories, 1989 10. Wiebe VJ, Benz CC, Shemano I, et al: Pharmacokinetics of toremifene and its metabolites in patients with advanced breast cancer. Cancer Chemother Pharmacol 25:247-251, 1990 11. Legha SS, Powell K, Buzdar AU, et al: Tamoxifeninduced hypercalcemia in breast cancer. Cancer 47:28032806, 1981 12. Enck RE, Rios CN: Tamoxifen treatment of metastatic breast cancer and antithrombin III levels. Cancer 53:2607-2609, 1984 13. Valvarra R, Pyrhonen S, Heikkinen M, et al: Toremifene, a new antiestrogenic compound for treatment of advanced breast cancer. Phase II study. Eur J Cancer Clin Oncol 24:785-790, 1988
Downloaded from ascopubs.org by RB DRAUGHTON LIBRARY on April 27, 2019 from 131.204.073.184 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
