Fax: (713) 794-4385
A phase 1 study of weekly everolimus (RAD001) in combination with docetaxel in patients with metastatic breast cancer
Article first published online: 17 OCT 2011
Copyright © 2011 American Cancer Society
Volume 118, Issue 9, pages 2378–2384, 1 May 2012
How to Cite
Moulder, S., Gladish, G., Ensor, J., Gonzalez-Angulo, A. M., Cristofanilli, M., Murray, J. L., Booser, D., Giordano, S. H., Brewster, A., Moore, J., Rivera, E., Hortobagyi, G. N. and Tran, H. T. (2012), A phase 1 study of weekly everolimus (RAD001) in combination with docetaxel in patients with metastatic breast cancer. Cancer, 118: 2378–2384. doi: 10.1002/cncr.26571
Fax: (713) 794-4385
- Issue published online: 20 APR 2012
- Article first published online: 17 OCT 2011
- Manuscript Accepted: 15 AUG 2011
- Manuscript Revised: 1 AUG 2011
- Manuscript Received: 2 MAY 2011
- phase 1;
- metastatic breast cancer
Inhibition of mammalian target of rapamycin with everolimus may improve the efficacy of taxanes. Everolimus and docetaxel are both metabolized by CYP3A4, which could result in a pharmacokinetic (PK) interaction.
Fifteen patients with metastatic breast cancer were treated with docetaxel (doses of 40-75 mg/m2 intravenously on day 1 of a 21-day cycle) in combination with everolimus (doses ranging from 20 to 50 mg orally on days 1 and 8 of a 21-day cycle) in a phase 1 trial using the continuous reassessment method to determine maximum tolerated dose. The first 2 patients developed a dose-limiting toxicity (neutropenic infection), prompting a mandatory dose reduction and PK evaluation of both everolimus and docetaxel for patients enrolled in subsequent dosing cohorts.
Fifteen patients were treated. Dose-limiting toxicity included grade 3 mucositis (n = 1), prolonged grade 4 neutropenia (n = 1), and grade 3 infection/febrile neutropenia (n = 3). Day 8 of everolimus was commonly held for neutropenia despite a dose reduction in docetaxel to 40 mg/m2. Eleven patients underwent complete PK evaluation for everolimus, and 9 patients underwent complete PK evaluation for both everolimus and docetaxel. Widely variable changes in clearance were seen for both drugs, and the study was terminated because of lack of efficacy and concerns regarding toxicity seen with the combination.
Weekly everolimus in combination with docetaxel every 3 weeks was associated with excessive neutropenia and variable clearance of both drugs, making combination therapy unpredictable, even at low doses of both drugs. Cancer 2012. © 2011 American Cancer Society.
Taxanes affect the formation of microtubules, protein polymers essential for maintenance of cytoskeleton organization and mitotic spindle formation. Taxanes have demonstrated activity in a variety of cancer types and are currently approved for the treatment of localized and advanced breast cancer. Docetaxel has been compared with doxorubicin and was found to significantly improve response rate (47.8% vs 33.3%; P = .008) in chemotherapy naive patients with advanced breast cancer.1 Although not statistically significant, there was a trend in favor of docetaxel for improvement in time to progression, although no improvement was seen in overall survival (OS). Docetaxel has also demonstrated response rates of 30% to 42% and time-to-treatment failure of 4.2 to 6.5 months in patients with advanced breast cancer previously treated with anthracyclines.2-6 Docetaxel has also been compared with paclitaxel given on a triweekly schedule and demonstrated significantly superior time to progression (5.7 vs 3.6 months, P < .001) and OS (15.4 vs 12.7 months, P = 0.03).3
Activation of the phosphatidylinositol 3-kinase (PI3K)-Akt signaling pathway is intricately involved in cancer cell growth and survival.7 The serine/threonine kinase mammalian target of rapamycin (mTOR) is a downstream effector protein that is activated by Akt and has been implicated in cancer cell resistance to drugs that inhibit microtubule function, such as taxanes.8, 9 In preclinical models, taxane therapy induced the activation of mTOR, and pretreatment with mTOR inhibitors, such as rapamycin or everolimus, increased taxane-induced cell death.8, 10
Everolimus is a novel macrolide that inhibits mTOR by binding with high affinity to its intracellular receptor FKBP12. The resulting complex binds to mTOR, which prevents downstream signaling.11 Everolimus has been developed as both an immunosuppressant for the treatment of transplant organ rejection and as an anticancer agent. A phase 1 study in solid tumors demonstrated dose-limiting toxicity (DLT) of mucositis and fatigue when everolimus was administered as a single agent using a weekly schedule. The drug's terminal half-life was 30 hours (range, 26-38 hours), and the area under the curve increased proportional to the dose.11
Everolimus and docetaxel are both extensively metabolized by the cytochrome CYP3A4 isoenzyme.12, 13 Docetaxel is metabolized into a primary metabolite, M2, through oxidation of the tert-butyl ester side group by CYP3A4.13 The metabolites of docetaxel are exceedingly less active than the parent compound; thus, given the drug's narrow therapeutic window, variations in the activity of CYP3A4 could result in substantial differences in toxicity if given in combination with inhibitors of CYP3A4. Based upon these data, we performed a phase 1 study of everolimus in combination with docetaxel for the treatment of patients with advanced breast cancer. The purpose was to establish the maximum tolerated dose of the combination, determine the DLTs, and identify antitumor activity.
MATERIALS AND METHODS
Patients were ≥18 years of age, with metastatic breast cancer refractory to standard therapy, with adequate bone marrow (absolute neutrophil count [ANC], ≥1.5 × 109/L; platelets, ≥100 × 109/L; hemoglobin, ≥10g/dL), renal function (creatinine, ≤1.5 × upper limits of normal [ULN]), and hepatic function (bilirubin, ≤ULN; alkaline phosphatase, ≤5 × ULN; if alkaline phosphatase was ≤2.5 × ULN, alanine aminotransferase [ALT]/aspartate aminotransferase[AST] must have been ≤2.0 × ULN; if alkaline phosphatase was >2.5 but ≤5 × ULN, ALT/AST must have been ≤1.5 × ULN), and performance status 0 to 2 using the World Health Organization scale. Exclusion criteria included: thromboembolism within the prior 6 months or active thrombophlebitis, grade >2 neuropathy, primary central nervous system (CNS) tumors or known CNS metastases, uncontrolled infection, known history of human immunodeficiency virus seropositivity, active bleeding diathesis or anticoagulation with warfarin (except patients receiving 1 mg warfarin to prevent central venous catheter thrombosis), other concurrent severe and/or uncontrolled medical disease that could compromise participation in the study, impairment of gastrointestinal function or gastrointestinal disease that may significantly alter the absorption of everolimus, prior treatment with mitomycin C or nitrosoureas, surgery within 2 weeks before study entry, and concomitant therapy with strong inhibitors or inducers of the CYP3A4 isoenzyme.
Evaluation of DLT
Toxicity was defined using the National Cancer Institute Common Toxicity Criteria version 2.0. DLT was defined as any grade 4 nonhematological toxicity, grade 3 nonhematological toxicity unresponsive to supportive care measures, grade 4 neutropenia lasting >7 days, febrile neutropenia, or grade ≥2 neutropenia (ANC, <1.5 × 109/L) or thrombocytopenia (platelets, <75 × 109/L) that failed to revert to grade ≤1 (level at which treatment is permitted) by the time of the scheduled start of cycle 2. DLT also had to occur during the first cycle of therapy and be considered drug related by the treating physician.
Dose-finding in this trial was conducted using a continuous reassessment model. This approach models the dose-toxicity relation using a 1-parameter curve. A prior distribution is placed on the parameter before the trial. After each set of patients is treated and toxicity is observed, the distribution of the parameter is updated and the next dose level is selected based on the predicted toxicity.14
Drug Dosages and Schedules
Patients received dexamethasone 8 mg by mouth twice daily for 3 days beginning 24 hours before docetaxel administration. Docetaxel was administered as a 1-hour infusion on day 1 of a 21-day cycle (±48 hours). Everolimus was taken orally on days 1 (within 5 minutes of starting docetaxel infusion) and 8 of each 21-day cycle, except cycle 2 (Fig. 1). During cycle 2, everolimus was only administered on day 8. This schedule allowed for pharmacokinetic (PK) analysis of each drug as a single agent and as part of combination therapy. In the initial dosing cohorts, docetaxel was fixed at a dose of 75 mg/m2 with escalating doses of everolimus (Table 1); however, 2 of 2 patients developed DLT (infection/febrile neutropenia) in the first dosing cohort (everolimus dose = 30 mg on days 1 and 8). Neither of these patients underwent optional PK testing as defined within the protocol. The study was redesigned with mandatory PK sampling and used a 3-patient run-in dosing cohort of docetaxel (60 mg/m2 intravenously [IV] every 21 days) in combination with 20 mg everolimus on days 1 and 8. The PK analyses from these patients were used to determine dosing for subsequent cohorts (Table 1).
|Phase/Dose Level||Docetaxel Dose||Everolimus Dose||PK Sampling||No.||DLT|
|Dose level 1||75 mg/m2 IV Q3wk||30 mg PO weekly||No||2||2 febrile/infection neutropenia|
|Dose level 2||60 mg/m2 IV Q3wk||20 mg PO weekly||Yes (n = 2)||3||1 prolonged grade 4 neutropenia|
|Dose level 3||40 mg/m2 IV Q3wk||30 mg PO weekly||Yes (n = 2)||2||None|
|Dose level 4||40 mg/m2 IV Q3wk||50 mg PO weekly||Yes (n =6)||8||1 stomatitis, 1 febrile neutropenia|
Treatment-related toxicity had to resolve to grade ≤1 before administration of subsequent cycles of therapy. Day 8 of everolimus was held for grade ≥3 hematologic or nonhematologic toxicity. Patients with grade 4 hematologic toxicity or grade 3 or 4 nonhematologic toxicity underwent a 25% reduction in the dose of docetaxel during subsequent cycles. Patients who did not return to grade ≤1 toxicity within 2 weeks of the start of the next cycle of therapy or who developed recurrence of grade 4 hematologic or grade 3 to 4 nonhematologic toxicity despite a dose reduction in docetaxel were withdrawn from protocol therapy.
After the first 2 patients treated developed DLT, PK analyses were mandatory for the remaining patients enrolled. PK blood sampling for everolimus was performed on cycle 1 day 1 (in combination with docetaxel) and cycle 2 day 8 (everolimus alone) predose and at 1, 2, 5, 8, and 24 hours (±5 minutes) after administration. PK sampling for docetaxel was performed on cycle 1 day 1 (in combination with everolimus) and cycle 2 day 1 (docetaxel alone) using the same sampling schedule as for everolimus, beginning at the start of docetaxel infusion. Blood was collected by direct venipuncture or an indwelling cannula inserted in a vein not receiving chemotherapy infusion. Blood samples were centrifuged at 1500 rpm for 10 minutes at 5°C, and plasma samples were transferred into appropriately labeled vials and stored frozen (−80°C) until analysis. All drug analysis for both everolimus and docetaxel was completed at a Novartis (East Hanover, NJ) laboratory using validated assays (on file at Novartis). PK parameters for everolimus and docetaxel were estimated using standard noncompartmental methods (WinNonlin version 5.0; Pharsight Corporation, Mountain View, Calif).
Fifteen patients were treated in this trial. The first 2 patients treated at the initial dosing cohort (docetaxel 75 mg/m2 IV on day 1, everolimus 30 mg orally [PO] on days 1 and 8 of a 21-day cycle) had DLT (infection/fever with neutropenia), and subsequent cohorts underwent mandatory PK testing. Patient characteristics are listed in Table 2. Patients enrolled were heavily pretreated with chemotherapy. The median number of prior chemotherapy treatments, including chemotherapy administered in the adjuvant or neoadjuvant setting, was 4 (range, 2-7). All patients had received at least 1 line of chemotherapy in the metastatic setting, with the majority of patients receiving >2 chemotherapy regimens for metastatic disease (Tables 2 and 3).
|Age (median, 58 years; range, 44-73)|
|WHO performance status|
|Hormone receptor status|
|Disease sites (all lesions)|
|Prior chemotherapy for MBC|
|Pt #||Prior Therapy, No.a||Cycle 1 D Dose, mg/m2||D Cl With E||D AUC With E||DLT||Received C1D8 E||Cycle 2 D Dose, mg/m2||D Cl Alone||D AUC Alone||% Change D Clearanceb||Dose E, mg||% Change Clearance Eb|
After the initial 2 patients treated developed DLT, a 3-patient run-in (docetaxel 60 mg/m2 IV on day 1, everolimus 20 mg PO on days 1 and 8 of a 21-day cycle) with mandatory PK sampling demonstrated a significant decrease of docetaxel clearance of 42% in 1 patient and a slight increase (+29%) in another (Table 3). Because of a shipping error, cycle 2 day 1 blood samples were unavailable for analysis in the third patient treated within this cohort. In addition, all patients treated in this dosing cohort were unable to receive cycle 1 day 8 everolimus because of grade 3 to 4 neutropenia. Based upon these data, subsequent patients were treated with a reduced dose of docetaxel at 40 mg/m2 IV every 21 days with alternating doses of weekly everolimus.
DLT occurred in 2 of 2 patients (100%, both neutropenic fever/infection) treated at dose level 1 (75 mg/m2 docetaxel with 30 mg everolimus), in 1 of 3 patients (33%, prolonged grade 4 neutropenia) treated at dose level −1 (60 mg/m2 docetaxel with 20 mg everolimus), and 2 of 8 patients (25%, 1 grade 3 stomatitis and 1 febrile neutropenia) treated at dose level −2 (40 mg/m2 docetaxel with 50 mg everolimus), as listed in Table 3. No patients treated in dosing cohort −1 received day cycle 1 day 8 everolimus because of neutropenia, and 50% of patients treated in dosing cohort −2 also did not receive cycle 1 day 8 everolimus. Several patients who received everolimus on day 8 of cycle 1 in dosing cohorts −2 and −3 were unable to receive day 8 of therapy on subsequent cycles because of neutropenia. Two patients were treated in dosing cohort −3 (40 mg/m2 docetaxel with 30 mg everolimus) without DLT.
Myelosuppression was commonly seen, with the majority of patients (73%) developing grade 3-4 neutropenia (Table 4). Anemia and thrombocytopenia were less common and were low grade (Table 4). Commonly seen nonhematologic toxicities are listed in Table 3 and include low-grade fatigue (40%), mucositis (20%), and hyperglycemia (20%). One patient had grade 3 mucositis that was persistent and considered DLT. Other grade 2 nonhematologic toxicities occurred in only 7% of patients and included gastritis, hypoalbuminemia, nasal discharge, myalgia, rash, and vomiting.
|Toxicity||Grade 2||Grade 3||Grade 4|
|Febrile neutropenia||0||2 (13%)||0|
|Leukopenia||3 (20%)||7 (47%)||2 (13%)|
|Lymphopenia||3 (20%)||2 (13%)||0|
|Neutropenia||1 (7%)||5 (33%)||6 (40%)|
|Non-neutropenia||2 (14%)||2 (14%)||0|
|Mucositis||3 (20%)||1 (7%)||0|
Of the 15 patients evaluable for response, none developed a partial or complete response to therapy (Table 5). Eight patients (53%) had stable disease (SD) as best response, and 2 of these patients had SD ≥6 months, for a clinical benefit rate of 13%. Seven patients (47%) developed progression of disease as best response to therapy.
|Patient #||Dose D (mg/m2)||Dose E (mg)||Response|
|5||60||20||SD, 6 cycles|
|6||40||30||SD, 8 cycles|
|7||40||50||SD, 4 cycles|
|10||40||50||SD, 6 cycles|
|11||40||50||SD, 4 cycles|
|12||40||50||SD, 8 cycles|
|13||40||50||SD, 6 cycles|
|15||40||50||SD, 6 cycles|
Of the 15 patients treated, PK data for docetaxel were available in 10 patients and for everolimus in 11 patients (the first 2 patients enrolled did not undergo optional PK testing, 3 patients had samples incorrectly processed after blood collection [docetaxel = 2, everolimus = 1], and 1 patient progressed after the first cycle of therapy). There was wide variation in the clearance of both drugs as listed in Table 3. Of the 13 patients treated during the mandatory PK requirements, 3 experienced DLT, and 7 did not receive cycle 1 day 8 dosing of everolimus. Of the 3 patients with DLT, 2 were noted to have inhibition of the clearance of everolimus by 9% to 23%, and 2 were noted to have increased docetaxel clearance by 9% to 26%. One patient had an increased clearance of both docetaxel and everolimus. Because of small numbers, no distinct correlation could be made with drug clearance and DLT or cycle 1 day 8 dosing of everolimus.
Several trials have investigated the role of mTOR inhibition with everolimus for the reversal of drug resistance in breast cancer. The aromatase inhibitor letrozole was combined with everolimus in the neoadjuvant setting and demonstrated improved clinical response rates with decreased proliferation (as measured by Ki67), compared with letrozole alone.15 Everolimus in combination with trastuzumab demonstrated a 15% response rate and a 34% clinical benefit rate in patients with human epidermal growth receptor 2 (HER2)+ breast cancer that had progressed on prior HER2-based therapy.16 Everolimus has also been combined with chemotherapy and trastuzumab for the treatment of HER2+, trastuzumab-refractory breast cancer, with notable response rates of 15% to 20%. A second study combining everolimus with paclitaxel and weekly trastuzumab demonstrated a response rate of 20%.17
Based upon preclinical data supporting the role of the PI3K/mTOR pathway in chemotherapy resistance, this study was designed to determine the maximum tolerated dose of the mTOR inhibitor everolimus given in combination with docetaxel. Unfortunately, this study demonstrated that combination dosing was associated with a high rate of neutropenia that often precluded sustained weekly dosing of everolimus. DLT of neutropenic infection/fever occurred in the first 2 patients treated with standard-dose docetaxel (75 mg/m2 IV every 3 weeks) and low-dose everolimus (30 mg PO weekly). Subsequent dosing cohorts using lower doses of docetaxel (40-60 mg/m2 IV every 3 weeks) did not substantially improve the feasibility of weekly dosing, as only half of the patients treated with 40 mg/m2 docetaxel in combination with 50 mg weekly everolimus were able to receive the cycle 1 day 8 everolimus dose.
Both everolimus and docetaxel are metabolized by the cytochrome CYP3A4 isoenzyme.12, 13 Metabolism of docetaxel to M2 reduces drug activity and primarily occurs by CYP3A4 oxidation of the tert-butyl side group of the parent compound.13 PK alterations have been noted when docetaxel was combined with either doxorubicin or etoposide.18 In addition, inhibition of CYP3A4 with ketoconazole resulted in a 49% decrease in clearance of docetaxel.19 Because docetaxel has a narrow therapeutic window, disruptions in drug metabolism could lead to the toxicity seen within the first dosing cohort. Noting this, subsequent patients treated after the first dosing cohort were required to undergo mandatory blood draws for PK analyses, which revealed a wide variation in metabolism of both docetaxel and everolimus (Table 3). The noted variations, however, were not strongly associated with DLT or withholding cycle 1 day 8 everolimus and thus cannot fully account for the difficulty maintaining an adequate dosing schedule. As previously mentioned, patients in this study were heavily pretreated with chemotherapy, which is also likely to confound the toxicity results.
No responses were noted on this study; however, 2 patients (13%) were noted to have stable disease for ≥6 months. The relevance of this observation is difficult to assess in the phase 1 setting, which may select for patients with slowly progressing or indolent disease.
Although there were 2 patients treated on dose level −3 (40 mg/m2 docetaxel with everolimus 30 mg PO weekly) who did not develop DLT or miss cycle 1 day 8 everolimus, this was felt to be substandard dosing for both drugs, and accrual to the trial was terminated because of lack of response and concerns over the regimen's feasibility. In addition, everolimus has been successfully administered in combination with paclitaxel, which is not extensively metabolized through CYP3A4; thus, subsequent research can be pursued with this combination.20 Because neutropenia was the predominant DLT, alternative dosing schedules may also improve the feasibility of this combination. For example, weekly dosing of docetaxel has been associated with less neutropenia and may be a viable alternative for solid tumors other than breast cancer, as weekly administration of docetaxel has demonstrated reduced efficacy compared with triweekly dosing in patients with metastatic breast cancer.21-23 Daily dosing of everolimus is also feasible and has been safely combined with every 3-week docetaxel (60 mg/m2) for the treatment of patients with advanced lung cancer.24
Supported in part by National Cancer Institute grant P30 CA016672.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 5Docetaxel compared with sequential methotrexate and 5-fluorouracil in patients with advanced breast cancer after anthracycline failure: a randomised phase III study with crossover on progression by the Scandinavian Breast Group. Eur J Cancer. 1999; 35: 1194-1201., , , et al.
- 17Everolimus in combination with weekly paclitaxel and trastuzumab in patients (pts) with HER2-overexpressing metastatic breast cancer (MBC) with prior resistance to trastuzumab and taxanes: a multicenter phase II clinical trial. J Clin Oncol. 2010; 28(suppl; abstract 1013)., , , et al.