A phase 2 study of the oral mammalian target of rapamycin inhibitor, everolimus, in patients with recurrent endometrial carcinoma

Abstract

BACKGROUND:

Dysregulation of phosphatase and tensin homolog (PTEN) and the gene that encodes the p110α catalytic subunit of phosphatidylinositol-3-kinase (PI3K), PIK3CA, are the most common mutations in endometrial carcinoma (EC). Loss of PTEN or activation of PIK3CA results in constitutive activation of AKT, which leads to up-regulation of mammalian target of rapamycin (mTOR). Everolimus is an oral rapamycin analog that acts by selectively inhibiting mTOR.

METHODS:

A single-institution, open-labeled, phase 2 study of everolimus in patients with measurable recurrent EC who had failed at least 1 and no more than 2 prior chemotherapeutic regimens was performed. Everolimus was administered at a dose of 10 mg orally daily for 28-day cycles. Patients were treated until disease progression or toxicity. The primary endpoint was clinical benefit response (CBR), defined as a confirmed complete or partial response or prolonged stable disease (SD) (≥8 weeks). Inclusion was limited to patients with endometrioid histology.

RESULTS:

A total of 35 patients were enrolled (median age, 58 years; range, 38-81 years). A total of 81 cycles were administered. Twelve of 28 (43%) evaluable patients had not developed disease progression at the time of the first objective evaluation (8 weeks). All these patients had SD (median, 4.5 cycles; range, 2-10 cycles). Six of the 28 (21%) patients had a confirmed CBR at 20 weeks of therapy. Patients with CBR discontinued treatment because of toxicity (6 patients), disease progression (5 patients), and noncompliance (1 patient). Seven patients were unevaluable after receiving ≤1 cycle because of toxicity (5 patients) or noncompliance (2 patients). The most common drug-related toxicities were fatigue, anemia, pain, lymphopenia, and nausea.

CONCLUSIONS:

Everolimus demonstrated encouraging single-agent CBR in pretreated patients with recurrent endometrioid EC. Future studies will evaluate this agent in combination with hormonal and/or cytotoxic therapy. Cancer 2010. © 2010 American Cancer Society.

In the United States, endometrial cancer remains the most commonly diagnosed gynecologic malignancy. Although the majority of women with incident cases will be cured with surgery alone or surgery in combination with adjuvant therapy, >8000 women die annually, predominately as a result of acquired resistance to conventional therapy. This observation has fueled intense investigation into alternative strategies, particularly those able to leverage an increasing knowledge of the molecular regulation of metastasis as well as the proliferation and survival of tumor cells.

One such axis with broad appeal in many solid tumors, including endometrial cancer, is the phosphatidylinositol-3-kinase (PI3K)/Akt pathway. Components of this pathway and its regulators are frequently constitutively activated or mutated, promoting carcinogenesis. For example, phosphatase with tensin homology (PTEN) (which is located on chromosome 10), a tumor suppressor gene regulating activation of Akt from PI3K, is absent or inactivated by mutation in 40% to 80% of endometrial cancers and PIK3CA, the gene that encodes the p110α catalytic subunit of PI3K, can undergo gain-of-function mutation in 39% of endometrial cancers.1-4 Downstream effects include hyperactivation of Akt5 and, in turn, its downstream effector proteins including the mammalian target of rapamycin (mTOR). mTOR acts as a sensor that integrates extracellular and intracellular events, coordinating growth and proliferation. mTOR is an intracellular protein kinase implicated in cell cycle control and specifically in the progression of cells from the G₁ to S-phase.

Rapamycin and its derivatives inhibit the function of mTOR. Preclinical studies have revealed that mTOR inhibitors potently arrest the growth of cells derived from rhabdomyosarcoma; neuroblastoma; glioblastoma; small cell lung cancer; osteosarcoma; pancreatic, breast, and prostate cancers; melanoma; leukemia; and lymphoma.6

In endometrial cancer cell lines, rapamycin demonstrated a growth inhibitory effect through induction of cell cycle arrest,7 and rapamycin inhibited phosphorylation of downstream targets of mTOR (S6 kinase [S6K] and 4E-binding protein 1 [4E-BP1]). In PTEN heterozygote animal models, mTOR inhibition reportedly ydecreases the development and progression of endometrial hyperplastic lesions.8, 9 These effects are primarily the result of decreased cellular proliferation and increased apoptosis.

The purpose of the current study was to evaluate the benefit of everolimus (Afinitor; Novartis Pharma, Basel, Switzerland) in the treatment of women with previously treated endometrial cancer. The primary efficacy endpoint was clinical benefit response (CBR), defined as a confirmed complete or partial response or prolonged stable disease (SD) (≥8 weeks) according to Response Evaluation Criteria in Solid Tumors (RECIST) criteria.

MATERIALS AND METHODS

The current study was a phase 2, open-labeled trial conducted at The University of Texas M. D. Anderson Cancer Center. Everolimus was kindly provided by Novartis Pharma. This study is registered on the clinical trial Web site of the National Cancer Institute (available at: clinicaltrials.gov, NCT00087685). Institutional Review Board approval was obtained.

Patient Population

Patients with progressive or recurrent endometrial carcinoma who had received 1 or 2 prior chemotherapeutic regimens were entered into this investigator-initiated phase 2 trial. Eligible patients were required to have histologically confirmed endometrioid endometrial carcinoma. Patients with serous, clear cell, carcinosarcoma, sarcoma, or other histologic subtypes were excluded. Patients were required to have a Zubrod performance status of 0 to 2 and no history of an invasive malignancy other than endometrial cancer within a 5-year period before trial entry. Patients were required to have measurable disease (defined as lesions measuring ≥2 cm in at least 1 dimension when evaluated by physical examination or conventional radiographic imaging, or measuring ≥1 cm in at least 1 dimension when evaluated by spiral computed tomography [CT]). Ascites and pleural effusions were not considered measurable disease.

No prior treatment with an mTOR inhibitor was permitted, nor was concomitant anticancer treatment. Patients with active infections were excluded. Pretreatment hematologic, renal, and hepatic function tests were required to be grade 0 or 1 according to Common Terminology Criteria for Adverse Events (CTCAE) (version 3.0).

Pretreatment evaluation included a physical examination; assessment of Zubrod performance status; measurement of index lesion(s) for response assessment clinically or via CT scan, chest x-ray, or magnetic resonance imaging; and hematologic and biochemical blood tests. Other testing was performed at the discretion of the treating physician. Written informed consent was obtained from all patients before trial entry.

Treatment Plan and Response Evaluation

All patients were treated on an outpatient basis and were treated until disease progression, dose-limiting toxicity, or confirmed complete clinical response. All patients had treatment initiated with an oral dose of 10 mg daily. One dose reduction was allowed for ≥grade 2 hematologic or nonhematologic toxicity. One cycle of therapy was considered 4 weeks.

Records of study medication used, dosages administered, and intervals between visits were kept during the study. Drug accountability was monitored by the study coordinator during the study and at the completion of the trial.

The primary efficacy endpoint was CBR, defined as a complete or partial response or prolonged SD (≥8 weeks) by RECIST criteria. Response was defined according to RECIST criteria (version 1.0). Follow-up imaging for all patients not demonstrating progressive disease at the time of the initial clinical evaluation (8 weeks) was performed every 12 weeks thereafter. Patients were evaluated sooner if they demonstrated obvious disease progression by physical examination, clinical deterioration, or new symptomatology suggestive of clinical disease progression. We confirmed the previous best response to therapy using these subsequent imaging studies. As long as the tumor was stable or had decreased in size and the patient was clinically stable, treatment would continue. A secondary endpoint was to determine the nature and degree of toxicity of everolimus in this cohort of patients. Toxicity was assessed using the National Cancer Institute Common Toxicity Criteria (version 3.0). A ≥grade 2 toxicity required treatment to be withheld until recovery to a toxicity of at least grade 1. Response duration was measured from the time measurement criteria were first evaluated until disease progression.

Statistical Analysis

This study incorporated a Bayesian design with multiple early stopping points. The target CBR rate was 20%. CBR was evaluated as patients were accrued, and the trial was to be stopped if there was evidence that the target CBR rate could not be met; that is, given the outcomes of the patients who had already been evaluated, if we determined that there was a <10% chance that the CBR rate was ≥20%, the trial would have been stopped.

RESULTS

Patients

A total of 35 patients were enrolled in the trial between January 2004 and December 2007. The median patient age was 58 years (range, 38-81 years). Study subjects received a total of 81 cycles of therapy. The median time since prior chemotherapy was 3 months (range, 2-19 months). All 35 women were evaluable for toxicity; 28 were evaluable for response. Seven patients were unevaluable after receiving ≤1 cycle because of toxicity (5 patients) or noncompliance (2 patients). Baseline characteristics of the enrolled patients are listed in Table 1.

Table 1. Demographic Summary of Enrolled Patients (N = 35)

Characteristic	Value
Median age (range), y	58 (38-81)
Stage of disease
I	9
II	4
III	11
IV	6
Unstaged	5
Prior radiotherapy
Total	22
Pelvic ± vaginal brachytherapy	17
Vaginal brachytherapy alone	3
Palliative	2
No. of prior chemotherapy regimens
0	1
1	18
2	14
3	2
Prior hormonal/other therapy
Megestrol acetate	5
Aromatase inhibitor	2
Tamoxifen	1
Bevacizumab	1

Efficacy

Of the 28 evaluable patients, 12 (43%) had not demonstrated disease progression at the time of the first evaluation assessment. There were no complete or partial responses noted. The median duration of SD was 4.5 months (range, 2-10 months). Patients with SD discontinued treatment because of toxicity (6 patients), disease progression (5 patients), and noncompliance (1 patient). The confirmed CBR rate (imaging at 20 weeks of treatment) was 21% (6 of 28 patients). Of these 6 patients, 3 had treatment discontinued because of progressive disease, 2 patients developed treatment-related pulmonary toxicity, and 1 patient developed recurrent grade 3 neutropenia.

Safety

The safety population of 35 patients included all patients who received at least 1 dose of the study drug. Seventeen patients (49%) required a dose reduction to 5 mg daily because of stomatitis (10 patients); hyperglycemia (3 patients); or thrombocytopenia, hypertriglyceridemia, rash, and fever (1 patient each). The most common adverse events considered possibly, probably, or definitely related to everolimus were fatigue, anemia, pain, lymphopenia, and nausea (Table 2).

Table 2. Adverse Events Considered Possibly, Probably, or Definitely Related to Everolimusa

Event	Overall	Grades 3 to 4
	No. %	No. %
Anorexia	7 (20)	1 (3)
Constipation	6 (17)	1 (3)
Dyspnea	1 (3)	0 (0)
Fatigue	19 (54)	8 (23)
Headache	3 (9)	1 (3)
Mood changes	2 (5)	0 (0)
Mucositis	9 (26)	0 (0)
Muscle weakness	2 (5)	0 (0)
Nausea	12 (34)	4 (11)
Pain (all sites)	13 (37)	0 (0)
Pneumonitis	2 (5)	0 (0)
Pruritis	1 (3)	0 (0)
Sensory neuropathy	2 (5)	0 (0)
Skin rash	4 (11)	0 (0)
Sore throat	1 (3)	0 (0)
Taste alteration	1 (3)	0 (0)
Vomiting	5 (14)	1 (3)
Weight loss	1 (3)	0 (0)
Anemia	15 (43)	3 (9)
Leukopenia	7 (20)	2 (5)
Lymphopenia	13 (37)	10 (29)
Neutropenia	6 (17)	1 (3)
Thrombocytopenia	4 (11)	1 (3)
Alkaline phosphatase	3 (9)	0 (0)
Bilirubin	1 (3)	1 (3)
SGPT	2 (5)	0 (0)
SGOT	2 (5)	0 (0)
Hypercholesterol	6 (17)	1 (3)
Elevated glucose	8 (23)	3 (9)
Elevated magnesium	1 (3)	1 (3)
Hypokalemia	1 (3)	1 (3)
Hyponatremia	2 (5)	2 (5)
Hypertriglyceride	5 (14)	1 (3)

• a Adverse events were graded according to Common Toxicity Terminology Criteria for Adverse Events (CTCAE) (version 3.0).

DISCUSSION

The results of the current phase 2 study of everolimus for the treatment of patients with recurrent or persistent endometrioid endometrial cancer demonstrated a relatively high rate of SD with monotherapy using this oral agent.

Oza et al from the National Cancer Institute of Canada evaluated CCI-779 (temsirolimus; Wyeth Pharmaceuticals, Madison, NJ) in patients with untreated (as first-line treatment) metastatic or recurrent endometrial cancer.10 Temsirolimus is an mTOR inhibitor available in an intravenous formulation. In the study by Oza et al, there was a 25% partial response rate (median, 5.1 months) and a 57% SD rate (median, 8.7 months). There was no correlation noted between biologic markers (including PTEN) and response. Their population included patients with serous and clear cell histologies in addition to the endometrioid histology. At the 2008 American Society of Clinical Oncology annual meeting, the same group presented data obtained in pretreated patients, similar to the population in the current study. Two patients, or 7%, achieved a partial response and 44% had SD for a median of 3.5 months.11

Another mTOR inhibitor, deforolimus (AP23573), was also studied in a multisite trial of 45 patients with recurrent endometrial cancer by Colombo et al. These investigators found a 7% partial response rate and a 26% SD rate. In the majority of the responders, the duration of response was <4 months.12 See Table 3 for a summary of these data.10-12

Table 3. Summary of mTOR Inhibitors Studied in Endometrial Cancers

Agent	No.	Population	RR	CBR	Duration, Months
Temsirolimus10	19	No prior therapy	25%	82%	8.7
Temsirolimus11	27	1 prior regimen	7%	51%	3.5
Deforolimus12	45	Up to 2 prior regimens	7%	33%	<4
Everolimus (current study)	35	1 or 2 prior regimens	0%	43%	4.5

• mTOR indicates mammalian target of rapamycin; RR, response rate (complete response [CR] + partial response [PR]); CBR, clinical benefit rate (CR + PR + stable disease).

In the current study, patients with histologic subtypes other than endometrioid were excluded. The rational for these criteria were based on the presence of PTEN mutations primarily in the endometrioid tumors. To the best of our knowledge, the other studies of mTOR inhibition in patients with recurrent endometrial cancer included all histologic subtypes and clinical benefit has been demonstrated in patients with these other types.10, 11 We currently are conducting further translational studies in our group of patients to determine whether PTEN mutational status is predictive of response. Future studies of mTOR inhibition will likely not exclude patients with endometrial cancers based on histology.

The safety profile of everolimus was acceptable in the context of heavily pretreated patients with endometrial cancer. When the trial was initiated, the administration of lipid-lowering drugs for patients who developed ≥grade 2 hypertriglyceridemia or hypercholesterolemia was not allowed. Since the initiation of this trial, mTOR inhibition has been associated more strongly with these abnormalities. In the current study, approximately 17% of patients had elevated cholesterol and 14% had elevated triglycerides. Three patients with SD were removed because of these abnormalities. Future studies of everolimus should include a treatment modification to add a statin therapy in the setting of ≥grade 2 lipid abnormalities. Two patients with SD experienced pulmonary toxicity requiring withdrawal from the study. Both of these patients were receiving therapy for a prolonged period of time (6 months and 10 months, respectively). After discontinuation of treatment, toxicities resolved spontaneously. This pulmonary-related toxicity needs to be evaluated further in patients receiving prolonged mTOR inhibitor therapy.

Although the findings of the current study are encouraging and suggest activity similar to that found in the evaluation of other mTOR inhibitors, further investigation is needed to evaluate the role of mTOR inhibition in the treatment of endometrial cancer. Potential benefits of everolimus are its oral bioavailability and relatively favorable toxicity profile. In addition, although combination therapy may be a more attractive (and effective) alternative, it is unclear which type of therapy (ie, hormonal, cytotoxic, or radiation) should be combined with everolimus or other mTOR inhibitors.

The Gynecologic Oncology Group is initiating a phase 2 trial of carboplatin and paclitaxel in combination with temsirolimus as first-line therapy in patients with untreated recurrent or progressive endometrial cancer. The University of Texas M. D. Anderson Cancer Center initiated a phase 2 trial of everolimus in combination with letrozole for the treatment of patients with previously treated recurrent or progressive endometrial cancer. This combination has been evaluated as part of a neoadjuvant breast cancer trial.13 In that trial, everolimus was found to significantly increase the efficacy of letrozole in patients with estrogen receptor-positive breast cancer. The frequent expression of hormonal receptors in endometrial cancer, particularly in lower grade lesions, suggests a similar strategy may be relevant in this disease.

The statistical design of the current study incorporated SD at 8 weeks as minimal criteria for meeting CBR. All 12 patients who were considered to have SD underwent a radiological assessment at no earlier than 8 weeks. To confirm this single-agent activity, patients were reassessed after an additional 3 cycles (12 weeks) of therapy. Greater than 20% of patients in this cohort who were heavily pretreated demonstrated a confirmed CBR after 20 weeks of treatment.

In conclusion, mTOR inhibition with the oral agent everolimus in patients with recurrent endometrioid endometrial carcinoma appears to demonstrate clinical benefit. Although achieving disease remission while receiving single-agent therapy appears unlikely, limiting disease growth by interrupting this key biologic pathway is a reasonable approach for disease control. We currently have trials underway that will evaluate the role of combination therapy with aromatase inhibition as well as with cytotoxic therapy.

CONFLICT OF INTEREST DISCLOSURES

This research was supported in part by the Ann Rife Cox Chair in Gynecology, The University of Texas M. D. Anderson Cancer Center.