Drug Approval Summaries: Arsenic Trioxide, Tamoxifen Citrate, Anastrazole, Paclitaxel, Bexarotene

Abstract

This report summarizes information on drugs recently approved by the Food and Drug Administration, Office of Drug Evaluation I, Division of Oncology Drug Products. Five applications supporting new claims will be discussed: Trisenox^™ (arsenic trioxide) for induction of remission and consolidation in patients with acute promyelocytic leukemia who are refractory to, or have relapsed from, retinoid and anthracycline chemotherapy, and whose disease is characterized by the presence of the t(15;17) translocation or PML/RAR-alpha gene expression; Nolvadex^® (tamoxifen citrate) in women with ductal carcinoma in situ, following breast surgery and radiation, to reduce the risk of invasive breast cancer; Arimidex^® (anastrazole) for first-line treatment of postmenopausal women with hormone receptor positive or hormone receptor unknown locally advanced or metastatic breast cancer; Taxol^® (paclitaxel), 175 mg/m² by 3 h infusion in combination with cisplatin for first-line treatment of advanced ovarian cancer; and Targretin^® gel (bexarotene) for the topical treatment of cutaneous lesions in patients with stage IA and IB cutaneous T-cell lymphoma who have not tolerated other therapies or who have refractory or persistent disease.

Information provided includes rationale for drug development, study design, efficacy and safety results, and pertinent literature references.

Trisenox (arsenic trioxide)

Sponsor: Cell Therapeutics

Indication:

TRISENOX^™ is indicated for induction of remission and consolidation in patients with acute promyelocytic leukemia (APL) who are refractory to, or have relapsed from, retinoid and anthracycline chemotherapy and whose APL is characterized by the presence of the t(15;17) translocation or PML/RAR-alpha gene expression.

Arsenic preparations have been recognized for their potential beneficial effects throughout the history of therapeutics. Medical texts of the 17th Century formally incorporated arsenic-based preparations, and in 19th Century America, several arsenic compounds were registered in the pharmacopoeia. In 1909, Paul Ehrlich developed the arsenic-based preparation salvarsan (sometimes known as the “Magic Bullet,” or compound 606), which had relatively selective anti-spirochete effects. Organic arsenic preparations such as melarsoprol are used for the treatment of African trypanosomiasis involving the central nervous system and other protozoal infections [1].

In China, arsenic trioxide (As₂O₃) was introduced as a treatment for APL during the 1970s. Reported response rates ranged from 66% to 84%. Moreover, 28% (9/32) of the patients survived more than 10 years [2, 3].

APL is one of the acute myeloid leukemias (M3) classified by the French American British nomenclature. Genotypically, the cells have a characteristic translocation t(15:17) or, less commonly, t(11:17). The PML/RAR (promyelocytic leukemia/retinoic acid receptor) fusion protein, a result of the t(15:17) translocation, functions as an oncogene and blocks differentiation of the myelocytes at the promyelocyte stage. As summarized in recent reviews, current treatment practice is to use cytotoxic chemotherapy, typically an anthracycline ± cytarabine plus oral all-trans-retinoic acid (ATRA) to achieve remission, followed by further courses of chemotherapy for consolidation. An additional maintenance phase of oral ATRA or chemotherapy or both has been reported to improve long-term outcomes [4, 5].

About 20% to 30% of patients either do not achieve remission or relapse from combination ATRA and cytotoxic chemotherapy. Current practice is to retreat using ATRA with or without chemotherapy. Remission rates are variable and are presumed to be dependent upon time from the last previous dose of ATRA, with a longer time being more favorable. The evidence to support this impression has not been tested in randomized controlled studies [6].

A single-arm multicenter clinical study of 40 APL patients who relapsed after, or who were refractory to, prior chemotherapy and ATRA was submitted to support this application. Trisenox, 0.15 mg/kg/day intravenously over 1-2 h, was administered daily until the bone marrow was cleared of leukemia cells, up to a maximum of 60 days. Patients in complete remission were to receive consolidation with Trisenox for 25 additional doses for up to a five-week period. Consolidation began within three to six weeks of completion of induction.

Twenty-eight of the 40 Trisenox-treated patients (70%) had a complete response (CR). Median time to CR was 53 days (range 28 to 85 days). Three of five children, all older than five years, had a CR. Eighteen of 24 patients (75%) who had their last ATRA <1 year prior to Trisenox had a CR and 10 of 18 patients (56%) who had their last ATRA >1 year prior to Trisenox had a CR.

Following induction and consolidation, 16 patients received further Trisenox as maintenance therapy and 15 patients had bone marrow transplantation. At last follow-up, 27 of 40 patients were alive with a median follow-up time of 484 days (range 280-755), and 23 of 28 complete responders remained in complete response with a median follow-up time of 483 days (range 280-755).

Cardiac dysrhythmias have been reported with arsenic use. In this study, QT prolongation greater than 500 msec as measured by prolongation of the absolute QT or QT corrected interval (QTc) on an electrocardiogram was seen in 16 (38%) patients. One patient had a prolonged QTc of >500 msec which led to asymptomatic torsade de pointes during coadministration of amphotericin B. Prolongation of QT did not increase with repeated exposure to Trisenox. QT and QTc returned to baseline following cessation of arsenic trioxide.

The APL differentiation syndrome (fever, dyspnea, fluid retention, pleural effusions, pericardial effusions, leukocytosis) was observed in 9 patients (23%) and was severe in 3 patients (8%); none of the patients had to discontinue arsenic trioxide because of the APL differentiation syndrome.

Common adverse events, experienced by more than 40% of patients, were fatigue, fever, edema, nausea, anorexia, diarrhea, emesis, headache, insomnia, cough, dyspnea, dermatitis, tachycardia, pain, hypokalemia, hypomagnesemia, and hyperglycemia. The most common (>10%) Grade 3 or 4 adverse events were abdominal pain, epistaxis, dyspnea, hypoxia, bone pain, thrombocytopenia, neutropenia, hypokalemia, and hyperglycemia. Febrile neutropenia occurred in 3 patients (8%) and 2 patients (5%) had documented sepsis. A total of 2 patients discontinued arsenic trioxide and 3 patients died within 30 days of their last dose.

Trisenox remission induction therapy achieved a 70% complete response rate in APL patients who were refractory or relapsed after prior treatment with cytotoxic chemotherapy and ATRA. Durability of complete response could not be assessed, as many patients received maintenance therapies other than Trisenox. Investigators should be aware of the possible occurrence of cardiac arrhythimias and APL differentiation syndrome when prescribing this treatment.

Nolvadex^® (tamoxifen citrate)

Sponsor: Astra Zeneca Pharmaceuticals

Indication:

Nolvadex is indicated in women with ductal carcinoma in situ (DCIS), following breast surgery and radiation, to reduce the risk of invasive breast cancer.

Appropriate local therapy for DCIS includes lumpectomy followed by radiation therapy, an approach supported by the results of the National Surgical Adjuvant Breast Project (NSABP) B-17 trial [7-10]. In this study, 818 women with DCIS were randomized to undergo lumpectomy or lumpectomy followed by breast irradiation. After eight years of follow-up, in-breast recurrence occurred in 31% of women treated with surgery alone, compared to 13% of women treated with surgery and radiotherapy (p = 0.0001). Overall mortality was 1.6% at eight years irrespective of assigned treatment. Of the pathologic features examined in exploratory analyses as potential markers of local recurrence, only moderate to marked comedo necrosis was an independent risk factor. Tumor-free margins were not required for entry in this trial. Subsequent reports from NSABP and others have suggested that the adequacy of the resection margin, rather than comedo necrosis, is a better predictor of local recurrence [11].

Tamoxifen was initially approved by the Food and Drug Administration (FDA) in 1977 for treatment of metastatic breast cancer. Since then, tamoxifen has been approved to A) reduce recurrence and improve survival when given adjuvantly to node-positive and node-negative estrogen receptor positive (ER+) breast cancer patients; B) reduce the occurrence of contralateral breast cancer in women with breast cancer, and C) reduce the incidence of breast cancer in women at high risk for breast cancer [12].

Given tamoxifen's ability to decrease the incidence of breast cancer in high-risk women and in those with a history of a prior breast cancer, the NSABP initiated NSABP B-24. This study was designed to evaluate whether the addition of tamoxifen to lumpectomy and radiation therapy could reduce the incidence of ipsilateral and contralateral invasive breast cancer in women with DCIS [13]. Data from this trial were submitted to the Agency as an efficacy supplement for a new indication. NSABP B-24 was a prospective randomized double-blind placebo-controlled trial performed in 1,804 women with DCIS treated with lumpectomy and breast irradiation. Following lumpectomy, women were randomized to receive either tamoxifen 10 mg PO bid or placebo two pills daily for five years, started concomitantly with radiotherapy. Randomized study drug and radiotherapy were to begin within 56 days of lumpectomy.

The results of NSABP-24, as determined by the FDA, are shown in Table 1.

These results demonstrate a statistically significant reduction in the incidence of invasive breast cancer (the primary study endpoint) but not noninvasive breast cancer. The occurrence of all ipsilateral and all contralateral events was significantly decreased as well. The absolute difference in invasive breast cancer rates between treatment arms at five years was about 4%. Benefit could not be predicted by the baseline characteristics of the tumor or the patient. Overall survival was 97% in each arm after five years of follow-up.

Table 2 shows the serious adverse events as determined by the FDA.

The safety data were consistent with the known safety profile of tamoxifen. Tamoxifen was associated with an increased incidence of endometrial cancer, stroke, and thromboembolic events. There was no increased incidence of nonbreast, non-endometrial cancers. Data on myocardial events and eye events (corneal changes, color vision perception, retinal vein thrombosis, retinopathy, cataracts) were not prospectively collected.

These data demonstrate that tamoxifen significantly reduced the risk of developing an invasive breast cancer in women with DCIS treated with lumptectomy and radiation. Survival was similar in both treatment arms. Patients should be aware that the long-term, disease-free survival and overall survival after local therapy for DCIS are excellent. The decision regarding therapy with tamoxifen for reduction in breast cancer incidence should be based upon an individual assessment of the benefits and risks of tamoxifen therapy.

ARIMIDEX^® (anastrazole)

Sponsor: Astra Zeneca Pharmaceuticals

Indication:

Anastrazole is indicated for first-line treatment of postmenopausal women with hormone receptor positive or hormone receptor unknown locally advanced or metastatic breast cancer.

Arimidex is a selective nonsteroidal aromatase inhibitor. In postmenopausal women, estradiol is produced primarily from the conversion of androstenedione to estrone through the aromatase enzyme complex in peripheral tissues. Estrone is subsequently converted to estradiol. In postmenopausal women, Arimidex, at a daily dose of 1 mg, produced estradiol suppression of greater than 80%. Arimidex does not possess progestogenic, androgenic, or estrogenic activity [14-16].

Daily doses of Arimidex up to 10 mg do not effect cortisol or aldosterone secretion, measured before or after standard ACTH challenge testing. Supplemental glucocorticoids or mineralocorticoids are not needed.

Arimidex was initially approved for treatment of hormone receptor positive or hormone receptor unknown advanced breast cancer in postmenopausal women with disease progression following tamoxifen. Patients with ER-negative disease and patients who did not respond to previous tamoxifen therapy rarely respond to Arimidex.

Two randomized double-blind, double-dummy trials comparing the efficacy and safety of Arimidex 1 mg daily with tamoxifen 20 mg daily were submitted for this first-line application. The principal endpoints were time to progression and objective-response rate. Treatment tolerability was also compared and preliminary survival results were examined. A total of 1,021 postmenopausal women with advanced breast cancer who were hormone receptor positive or hormone receptor unknown were randomized to Arimidex (n = 511) or to tamoxifen (n = 510). Most patients had not received prior adjuvant therapy. A similar proportion in each treatment group had evidence of liver and other visceral involvement.

Tumor assessments were performed every four weeks for the first 24 weeks of trial treatment, every 12 weeks thereafter, and at the time of withdrawal (for any reason). Radiologic assessments were performed every 12 weeks.

Response rates (intent to treat population) are indicated in Table 3.

Time to progression (TTP) is indicated in Table 4. TTP appeared to be superior on the Arimidex arm in one of the two studies.

Survival data for Arimidex and tamoxifen are similar in both trials but are immature with 62% of patients censored. The sponsor agreed to submit annual updates of survival data.

A total of 415 (82.0%) women given Arimidex and 429 (84.0%) given tamoxifen reported at least one adverse event. The most frequently reported adverse event was hot flushes, which occurred in 131 (25.9%) subjects who were given Arimidex and 110 (21.5%) subjects who were given tamoxifen. Other commonly reported adverse events (≥10%) of subjects) were nausea, asthenia, pain, peripheral edema, and increased cough. Venous and/or arterial thromboembolic disease was reported in 23 Arimidex-treated women (4.5%) and 39 receiving tamoxifen (7.6%). Vaginal dryness or bleeding occurred in 20 Armidex-treated patients (4.0%) and 24 receiving tamoxifen (5.7%).

Arimidex was equivalent (non-inferior) to tamoxifen in terms of response rate. Time to progression favored Arimidex in one study and was non-inferior to tamoxifen in a second study. Both Arimidex and tamoxifen were well tolerated by most patients. These data indicate that Arimidex is an alternative first-line treatment for locally advanced or metastatic breast cancer in postmenopausal women with receptor positive or receptor unknown disease.

Taxol^® (paclitaxel)

Sponsor: Bristol-Myers Squibb

Indication:

For the first-line treatment of advanced ovarian cancer in combination with cisplatin. Approval is for the use of paclitaxel at a dose of 175 mg/m² administered intravenously over 3 h followed by cisplatin every three weeks.

Taxol was initially approved for first-line therapy of advanced carcinoma of the ovary as a 24-h infusion in combination with cisplatin. Approval was based on the results of a Gynecologic Oncology Group phase III trial (GOG-111). A total of 410 patients with stage III or IV disease (>1 cm of residual disease after staging laparotomy or distant metastases) were randomized to Taxol 135 mg/m² over 24 h and cisplatin 75 mg/m² or to cyclophosphamide 750 mg/m² and cisplatin 75 mg/m². Taxol- and cisplatin-treated patients had a significantly longer time to progression (median 16.6 versus 13.0 months, p = 0.0008) and survival (median 35.5 versus 24.2 months, p = 0.0002) [17].

The recent approval of Taxol by 3-h infusion for first-line therapy is based on European Organization for the Research and Treatment of Cancer (EORTC) CA139-209, a prospective multi-center, randomized trial of every-three-weekly TC (Taxol 175 mg/m² over 3 h and cisplatin 75 mg/m²) versus standard CC (cyclophosphamide 750 mg/m² and cisplatin 75 mg/m²) in patients with stage II-IV ovarian cancer. Optimally debulked (39% and 35%, respectively) and suboptimally debulked (61% and 65%, respectively) patients were entered.

TC-treated patients had a significant improvement in overall survival, time to progression, and clinical response rate (Table 5).

The results of this study of first-line therapy of advanced ovarian cancer study using a 3-h Taxol infusion schedule in combination with cisplatin are similar to those of the GOG-111 study using a 24-h Taxol infusion regimen. However, in the EORTC study approximately one-third of patients had “optimally” debulked disease whereas in the GOG study all patients had disease that had been “suboptimally” debulked.

In an exploratory analysis of the relative efficacy of the 3-h Taxol infusion and the 24-h infusion, the FDA compared outcomes in the suboptimally debulked patients treated in the EORTC and GOG trials. Results of this exploratory analysis are shown in Table 6.

In the EORTC trial, the hazard ratio in favor of Taxol treatment given by 3-h infusion was 0.73, which represents a 27% reduction in the risk of death. In the GOG trial, the hazard ratio favoring Taxol treatment given by 24 h infusion was 0.64, which represents a 36% reduction in the risk of death. Because of the cross-study comparisons and the width of the confidence intervals for the hazard ratios, a definitive statement about the relative efficacy of these regimens cannot be made.

The median dose intensity of Taxol was higher in the EORTC study than in the GOG-111 study, 58.7 mg/m²/week versus 41 mg/m²/week, respectively. The use of more dose-intense Taxol regimen did not result in significant reduction in the overall cisplatin dose intensity. Despite the higher Taxol dose intensity, the regimen was less myelosuppressive, and more patients were able to receive ≥90% of the planned dose of Taxol compared with GOG-111 (87% versus 52%).

The toxicity profile of the 3-h Taxol regimen in the EORTC study was consistent with previously documented adverse events of paclitaxel but with an increase in neurotoxicity. Patients on TC were more likely to complete the planned therapy, compared to those treated with CC. Compared with the GOG-111, neurotoxicity emerged as a significant toxicity as well as arthralgias/myalgias (Table 7). Neutropenia occurred less frequently in TC-treated patients than in the CC-treated patients and in all patients in the GOG trial, TC-treated and CC-treated.

Paclitaxel was approved as a 3-h infusion at a dose of 175 mg/m² in combination with cisplatin for first-line treatment of advanced-stage ovarian cancer. Patients and prescribing physicians should be aware of the degree of neurotoxicity associated with this infusion schedule which other studies have confirmed [18, 19].

Targretin^® Gel (bexarotene)

Sponsor: Ligand Pharmaceuticals

Indication:

Targretin^® (bexarotene) gel 1% is indicated for the topical treatment of cutaneous lesions in patients with stage IA and IB cutaneous T-cell lymphoma (CTCL) who have not tolerated other therapies or who have refractory or persistent disease after other therapies.

CTCL (mycosis fungoides) is typically a chronic, slowly progressive disease. About 1,000 new cases of CTCL are diagnosed annually in the U.S. The disease typically occurs in adults 40 to 60 years old. It occurs in all races with a 2:1 male:female ratio. Skin lesions may remain as patches or plaques (stage I) for many years before development of cutaneous tumors or visceral disease (stages II-IV) [20].

Therapy for CTCL is frequently given topically, especially in the earlier stages of the disease. Topical glucocorticoids, nitrogen mustard (mechlorethamine), carmustine (BCNU), psoralen plus ultraviolet-A radiation (PUVA), and electron beam radiation therapy (EBT) can induce remissions but do not alter the patient's long-term prognosis. For patients with advanced relapsing or nonresponsive disease, approved therapies include photopheresis (methoxsalen plus extracorporeal long-wave UV radiation) of white blood cells, systemic mechlorethamine (Mustargen^®), and targretin (bexarotene) capsules [21].

Topical nitrogen mustard, topical BCNU, and radiation therapies (EBT and PUVA) carry significant epidermal carcinogenic risk. Environmental exposure of household contacts and health care workers to nitrogen mustard and BCNU is also a concern. Radiation therapies may induce skin aging changes, telangiectasia, edema, radiation dermatitis, permanent alopecia, and chronic blepharitis. The incidence of drug hypersensitivity is reported to occur in 35% to 58% of patients treated with topical nitrogen mustard and in 5% to 10% of patients treated with BCNU [21].

Newer active therapies include interferon-α and -γ, and various retinoids including isotretinoin (13-cis-retinoic acid), etretinate, and arotinoid. Retinoids play critical roles in normal development and physiology by modulating cell growth, division, reproduction, differentiation, and immune function. They are also capable of inhibiting cell growth, inducing differentiation, and inducing apoptosis (programmed cell death) in a variety of tumor cell lines [22-24].

Scandinavian investigators demonstrated the utility of retinoids as part of combination CTCL therapies in the mid-1980s. They reported complete clearing of skin lesions or partial remissions with isotretinoin, either alone or in combination with PUVA, systemic anticancer drugs, or glucocorticoids. In the U.S., oral isotretinoin was also found to produce complete or partial remissions of CTCL. Up to 75% of patients had some clinical response, usually after about eight weeks of therapy, with 9% having a complete response [25].

Retinoid effects appear to result from changes in gene expression mediated through specific intracellular receptors (IRs). There are two subfamilies of retinoid IRs: retinoic acid receptors (RARs) and the retinoid X receptors (RXRs). The RARs and RXRs each have three subtypes designated RARα, RARβ, RARγ and RXRα, RXRβ, and RXRγ, respectively. Each receptor subtype is thought to control both unique and overlapping target genes [24, 26].

Targretin drug substance is a synthetic retinoid that preferentially interacts with the RXR subtypes. In contrast, ATRA is a naturally occurring hormone that binds with high affinity only to the RAR subtypes, while 9-cis-retinoic acid (9-cis-RA) is a “pan-agonist” (i.e., it binds and activates all known retinoid receptors, including RAR and RXR families). The exact mechanism of action of targretin gel in the treatment of CTCL is unknown.

Targretin gel was evaluated for treatment of patients with early-stage CTCL in a multicenter, open-label clinical trial. Patients with stage IA, IB, and IIA were eligible. These disease stages are characterized by eczematous patches, papules, or plaques. In stage IA, lesions cover less than 10% of skin surface (T1) while in stage IB, they cover more than 10% of the skin surface (T2). Stages IA and IB have no clinically abnormal peripheral lymph nodes. Node biopsies, if performed, are negative for CTCL. Stage IIA disease has T1 or T2 skin lesions and has clinically abnormal peripheral lymph nodes. Node biopsies, if performed, are negative for CTCL. All of these stages have no tumors and no involvement of visceral organs.

Study patients had to be refractory to, intolerant of, or reached a response plateau for at least six months on at least two prior therapies from the following list: PUVA, UVB, EBT, photopheresis, interferon, systemic cytotoxic chemotherapy, topical nitrogen mustard, or topical carmustine (BCNU). At least one of these qualifying prior treatments must have been topical nitrogen mustard, topical carmustine or a phototherapy (UVB, PUVA, or EBT). Topical steroids and systemic retinoids did not qualify.

Targretin gel was applied to all lesions, although a maximum of five lesions served as index lesions for efficacy evaluation. Since there was only one concentration of study drug in the study, exposure adjustments were made only by adjustments to the frequency of application. All patients were started on a dosing frequency of every other day (qod). The frequency of application was escalated, at one-week intervals, to 1% qd, then 1% bid, then 1% tid, and then 1% qid, as tolerated.

Response rate, the primary study endpoint, was determined by composite assessment (CA). The CA was generated by a summation of the grades, for each index lesion, of erythema, scaling, plaque elevation, hypo- or hyperpigmentation, and area of involvement. The sponsor proposed that the study be deemed successful if the observed response rate (CCR + PR) was at least 20% and the lower bound of the 95% confidence interval (centered around the observed response rate) excluded the theoretical maximal spontaneous response rate of 5%. Physicians global assessment, an alternative method for determining response, was not used because global photographs, necessary for the FDA to confirm responses, were not obtained.

A single phase II study was carried out in 50 patients, 94% of them had stage I disease. The median age of study participants was 64 years. Eighty percent of patients were white. The median time since first clinical diagnosis of CTCL was 73.1 (range 2.0-278.5) months and the median time from first pathologic diagnosis was 69.7 (range 1.8-254.4) months.

Among the 50 patients there were 13 responders in 47 evaluable patients (28%: 95% CI 16%-43%). For the 13 responders, the median time to response was 87.5 days (range 36-154 days). The median time to complete response (n = 1) was 174 days. The median duration of response was not reached but will exceed 149 days.

The safety of Targretin gel has been assessed in clinical studies of 117 patients with CTCL who received Targretin gel for up to 172 weeks. In the multicenter open label study, 50 patients with CTCL received the medication for up to 98 weeks. The mean duration of therapy for these 50 patients was 199 days. The most common adverse events reported with an incidence at the application site of least 10% were rash, pruritus, skin disorder, and pain.

Adverse events leading to dose reduction or study drug discontinuation in at least two patients were rash, contact dermatitis, and pruritus.

Of the 49 patients (98%) who experienced any adverse event, most experienced events categorized as mild (9 patients, 18%) or moderate (27 patients, 54%). There were 12 patients (24%) who experienced at least one moderately severe adverse event. The most common moderately severe events were rash (7 patients, 14%) and pruritus (3 patients, 6%). Only one patient (2%) experienced a severe adverse event (rash).

In summary, targretin gel (1%) was approved for topical treatment of cutaneous lesions in patients with CTCL (stage IA and IB) who have not tolerated other therapies or who have refractory or persistent disease after prior therapies. Sufficient data were not provided to support the safety and efficacy of this product in stage II patients.

References