Two phase 2 trials of the novel Akt inhibitor perifosine in patients with advanced renal cell carcinoma after progression on vascular endothelial growth factor-targeted therapy

Authors

  • Daniel C. Cho MD,

    Corresponding author
    1. Division of Hematology and Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    • 330 Brookline Avenue, Boston, MA 02215; Fax: (617) 632-9260

    Search for more papers by this author
    • Fax: (617) 632-9260

  • Thomas E. Hutson DO,

    1. Division of Oncology, Baylor University Medical Center, Dallas, Texas
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Wolfram Samlowski MD,

    1. Comprehensive Cancer Centers of Nevada, Southern Hills, Nevada
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Peter Sportelli BS,

    1. Keryx Pharmaceuticals, Inc., New York, New York
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Brad Somer MD,

    1. The West Clinic, Memphis, Tennessee
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Paul Richards MD,

    1. Blue Ridge Cancer Care, Salem, Virginia
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Jeffrey A. Sosman MD,

    1. Division of Hematology and Oncology, Vanderbilt-Ingram Cancer Center, Nashville, Tennessee
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Igor Puzanov MD,

    1. Division of Hematology and Oncology, Vanderbilt-Ingram Cancer Center, Nashville, Tennessee
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • M. Dror Michaelson MD, PhD,

    1. Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Boston, Massachusetts
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Keith T. Flaherty MD,

    1. Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Boston, Massachusetts
    2. Division of Hematology and Oncology, University of Pennsylvania, Philadelphia, Pennsylvania
    3. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Robert A. Figlin MD,

    1. Division of Hematology and Oncology, University of Pennsylvania, Philadelphia, Pennsylvania
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author
  • Nicholas J. Vogelzang MD

    1. Cedars-Sinai Medical Center, Los Angeles, California
    2. Comprehensive Cancer Centers of Nevada, Las Vegas, Nevada
    Search for more papers by this author

Abstract

BACKGROUND:

The clinical activity of allosteric inhibitors of mammalian target of rapamycin (mTOR) inhibitors in renal cell carcinoma (RCC) may be limited by upstream activation of phosphatidylinositol 3 (PI3)-kinase/Akt resulting from mTOR1 inhibition. On the basis of this rationale, 2 independent phase 2 trials (Perifosine 228 and 231) were conducted to assess the efficacy and safety of the novel Akt inhibitor perifosine in patients with advanced RCC who had failed on previous vascular endothelial growth factor (VEGF)-targeted therapy.

METHODS:

In the Perifosine 228 trial, 24 patients with advanced RCC received oral perifosine (100 mg daily). Perifosine 231 enrolled 2 groups that received daily oral perifosine (100 mg daily): Group A comprised 32 patients who had received no prior mTOR inhibitor, and Group B comprised 18 patients who had received 1 prior mTOR inhibitor.

RESULTS:

In the Perifosine 228 trial, 1 patient achieved a partial response (objective response rate, 4%; 95% confidence interval, 0.7%-20%), and 11 patients (46%) had stable disease as their best response. The median progression-free survival was 14.2 weeks (95% confidence interval, 7.7-21.6 weeks). In the Perifosine 231 trial, 5 patients achieved a partial response (objective response rate, 10%; 95% confidence interval, 4.5%-22.2%) and 16 patients (32%) had stable disease as their best response. The median progression-free survival was 14 weeks (95% confidence interval, 12.9, 20.7 weeks). Overall, perifosine was well tolerated, and there were very few grade 3 and 4 events. The most common toxicities included nausea, diarrhea, musculoskeletal pain, and fatigue.

CONCLUSIONS:

Although perifosine demonstrated activity in patients with advanced RCC after failure on VEGF-targeted therapy, its activity was not superior to currently available second-line agents. Nonetheless, perifosine may be worthy of further study in RCC in combination with other currently available therapies. Cancer 2012. © 2012 American Cancer Society.

INTRODUCTION

The therapeutic relevance of the kinase mammalian target of rapamycin (mTOR) in renal cell carcinoma (RCC) has been established by the demonstration of clinical efficacy of allosteric mTOR inhibitors in patients with advanced RCC.1, 2 Two such agents, the rapalogues temsirolimus and everolimus, are now approved by the US Food and Drug Administration for the treatment of patients with advanced RCC.1, 2 Despite prolongation of progression-free survival (PFS), responses to these drugs occur in <5% of patients, and the median PFS is usually only 4 to 6 months. Efforts to improve on these first-generation mTOR inhibitors have focused primarily on identifying and opposing potential mechanisms of resistance to these agents.

Rapalogue therapy may induce the feedback activation of the phosphatidylinositol 3-kinase (PI3K)/Akt pathway in RCC tumor cells. Treatment with rapamycin in some patients has resulted in the activation of PI3K through a feedback loop involving the insulin-like growth factor-1 receptor (IGFR-1).3 It also is believed that rapalogues are more active in suppressing the function of mTOR complex 1 (TORC1), the complex that includes mTOR and regulatory-associated protein of TOR (raptor), and that they have less activity against TORC2, the complex that includes mTOR and rapamycin-insensitive companion of TOR (rictor). Treatment with rapamycin in some tumor cell types resulted in the derepression of TORC2, leading to the direct activation of Akt through enhanced TORC2-mediated phosphorylation of Akt on serine 473 (Ser473).4 The finding that PI3K/Akt activates numerous survival and proliferation-promoting kinases, transcription factors, and proteins other than mTOR has lead to the hypothesis that greater therapeutic results may be possible by circumventing the potential feedback loop and targeting PI3K or Akt upstream of mTOR.

Perifosine (KRX-0401; 1,1-dimethyl-4 [([octadecyloxy]hydroxyphosphinyl)oxy]-piperidinium inner salt) is a synthetic, substituted heterocyclic alkylphospholipid that has demonstrated the ability to inhibit Akt activity. Perifosine inhibits the activation of Akt by interfering with the interaction between the pleckstrin homology domain of Akt and phosphatidylinositol phosphate (PIP3), thereby precluding its translocation to the plasma membrane and activation through phosphorylation by pyruvate dehydrogenase kinase, isozyme 1 (PDK1).5 Furthermore, it was demonstrated recently that perifosine both induced autophagy and inhibited the assembly of the mTOR complexes by promoting the degradation of Akt, mTOR, rictor, raptor, and the p70 S6 ribosomal kinase through a glycogen synthase kinase-3/F-box and WD repeat domain-containing 7 (GSK3/FBW7)-dependent pathway.6

Perifosine has been tested in phase 1 and 2 trials using a variety of dosing schedules and has been well tolerated. Nausea and diarrhea have been the most frequent toxicities observed. In a recent randomized phase 2 trial comparing different perifosine dosing schedules (daily vs weekly administration) in patients with a variety of solid tumors, perifosine was active in patients with RCC (Perifosine trial 207).7 Activity was observed with both dosing schedules. Of the 212 patients enrolled, 13 had RCC, and 7 of these were evaluable for response. Three patients (43%) had documented partial responses (PRs), and an additional 2 patients (29%) had long-term disease stabilization. In addition, preclinical investigations revealed that perifosine reliably inhibited Akt and S6 ribosomal protein phosphorylation and induced apoptosis of RCC cell lines in vitro.8 On the basis of these promising early clinical and preclinical data, and to define its single-agent efficacy, perifosine was assessed in 2 independent phase 2 clinical trials (Perifosine 228 and Perifosine 231) in previously treated patients with advanced RCC.

MATERIALS AND METHODS

Inclusion Criteria

To be included in these 2 studies, adult patients (aged >18 years) with an eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 with confirmed metastatic RCC were required to have documented Response Evaluation Criteria in Solid Tumors (RECIST)-defined disease progression after treatment with sunitinib or sorafenib. Prior therapy with bevacizumab and/or cytokines (ie, interleukin-2, interferon) was permitted, as was prior vaccine therapy in the adjuvant setting. Patients enrolled in Perifosine 228 and in Group A of Perifosine 231 were not allowed to have received prior treatment with an mTOR inhibitor, whereas patients enrolled in Group B of Perifosine 231 were allowed to have failed therapy with 1 prior mTOR inhibitor. No prior brain metastasis were allowed on Perifosine 228, whereas patients with central nervous system metastases were allowed on Perifosine 231 provided that: 1) central nervous system disease was documented by stable or regressing lesions on magnetic resonance imaging (MRI) after radiation therapy, surgery, or both; 2) patients were off corticosteroids for at least 1 month; and 3) patients had completed radiation therapy >28 days before study entry.

Study Design

Both studies were single-arm, open-label phase 2 trials. Informed consent was obtained from all patients before any study-related activities. In both studies, after meeting eligibility requirements, patients received oral perifosine at a dose of 100 mg once daily. Perifosine had been studied previously in >1000 patients at both daily and weekly dosing schedules. The lower 100-mg daily dose was selected because of the tolerability and efficacy observed in those previous studies.9 In Perifosine 228, treatment cycles were 6 weeks in duration, but patients were evaluated by physical examination and safety laboratory evaluations every 3 weeks. In Perifosine 231, treatment cycles were 4 weeks in duration, and patients were evaluated by physical examination and safety laboratory evaluations on day 1 of each cycle. In both studies, patients were allowed to continue on study therapy provided that all nonhematologic toxicities had resolved to either grade 1 or tolerable grade 2 according to version 3 of the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE 3.0).

Dose modification for toxicity was based on the grade of the adverse event. For persistent grade 2 gastrointestinal adverse events, the dose was divided, and patients received 50 mg twice daily. A dose reduction to 50 mg once daily was allowed for persistent gastrointestinal side effects. Other grade 2 adverse events first were treated symptomatically without dose modification. If the adverse event persisted or the regimen remained unacceptable (intolerable) to the patient, then it was allowed to reduce the dose by 50 mg daily. Patients were allowed a drug holiday for up 2 weeks to allow for recovery from toxicity before the initiation of the reduced dose. For grade 3 or 4 toxicities that were attributable to perifosine, drug was held, and the patient was re-evaluated at least weekly until toxicity improved to grade ≤1 or baseline. If the toxicity resolved within 2 weeks, then the patient was allowed to restart perifosine at a reduced dose of 50 mg. Patients for whom perifosine-related toxicities did not resolve to grade ≤1 or baseline within 2 weeks or who could not tolerate a reduced dose of 50 mg once daily were removed from the study.

Before enrollment, all patients underwent a complete history and physical examination, a review of concomitant medications, ECOG performance status assessment, and documentation of evaluable disease. Pretreatment tests included a complete blood count, standard tests to document adequate organ function (eg, serum creatinine, liver function tests), prothrombin time/international normalized ratio, and radiographic evaluation of disease by either computed tomography (CT) scan or MRI. Radiographic studies were obtained ≤30 days before the first dose of drug. All other baseline evaluations were performed ≤14 days before the first dose of drug.

During the treatment phase of the study, patients on Perifosine 228 underwent a complete history and physical examination as well as routine blood testing every 3 weeks. Radiographic disease evaluations occurred every 6 weeks for the first 24 weeks, then every 12 weeks thereafter. On Perifosine 231, patients underwent a complete history and physical examination and routine blood testing every 4 weeks. Radiographic disease evaluations occurred every 12 weeks.

Outcome Measures

The primary objective for both protocols was to estimate the median PFS for all patients who received treatment with perifosine. PFS was measured from the date of registration until patients developed disease progression as defined by RECIST. The primary endpoint for both protocols was the proportion of progression-free patients at 12 weeks. Because the first scan in Perifosine 231 was at 12 weeks, the progression-free proportion at 12 weeks was calculated from first day of study drug treatment rather than from the time of registration. Patients who died without disease progression within 12 weeks and patients who began nonprotocol therapy without evidence of progression within 12 weeks were not considered to be progression-free at 12 weeks. Secondary efficacy outcomes included the objective response rate and the clinical benefit rate. The objective response rate was defined as the percentage of all treated patients who had a confirmed partial response (PR) or complete response (CR). All CRs and PRs required radiographic confirmation at 2 separate evaluations at least 28 days apart. Clinical benefit was defined as either an objective response according to RECIST or stable disease (SD) on the 12-week scan. Safety and toxicity were evaluated through physical examinations, laboratory evaluations, and adverse events, which were graded according to the NCI-CTCAE 3.0.

Statistical Considerations

The objective of both phase 2 single-arm trials was to estimate the PFS of tyrosine kinase inhibitor (TKI)-resistant patients with RCC who received treatment with perifosine. In the Perifosine 228 trial, sample size calculation was based on a binomial design, because all patients were to be observed for at least 12 weeks. The study hypothesis was that, for patients who progressed on sorafenib or sunitinib, the median PFS would be 6 weeks (ie, the next CT scan) and that observation of 10-week PFS during perifosine treatment would indicate that perifosine was worthy of further study. Assuming an exponential distribution, this would be equivalent to 44% of patients remaining alive and progression free at 12 weeks on perifosine compared with the a null value of 25%. To detect a 19% absolute improvement in the proportion that remained alive and progression free at 12 weeks with 90% power using a 1-sided test with a Type I error rate of 0.1, 43 patients were required. To allow for a 10% ineligibility rate, a goal accrual of 48 patients was chosen. Perifosine was to be declared worthy of further study if ≥14 patients remained alive and progression free at 12 weeks.

For the Perifosine 231 trial, 2 patient cohorts were studied: those who had previously failed on a vascular endothelial growth factor (VEGF) receptor inhibitor but not on an mTOR inhibitor (Group A) and those who had failed both on a VEGF receptor inhibitor and on an mTOR inhibitor (Group B). In Group A, perifosine was to be deemed worthy of further study if ≥35% of the patients responded or were progression free at 12 weeks, but it was not to be considered for further study if the rate was ≤15%. A 2-stage Simon design was used with 90% power using a 1-sided test with a Type I error rate of 0.05. In the first stage, 19 patients were to be entered. If ≤3 of the 19 patients responded or were progression free at 12 weeks, then accrual to the study was to be halted. Otherwise, a total of 44 patients were to be accrued, and perifosine would be considered for further study if ≥8 patients respond or were progression free at 12 weeks.

In Group B, perifosine was to be deemed worthy of further study if ≥25% of the patients responded or were progression free at 12 weeks, but it was not to be considered for further study if the rate was ≤10%. A 2-stage Simon design was also used for this group with a 1-sided test, a Type I error rate of 0.05, and a power of 80%. In the first stage, 18 patients were to be entered. If ≤2 patients responded or were progression free at 12 weeks, then accrual to the study was to be halted. Otherwise, a total of 43 patients were to be accrued, and perifosine would be considered for further study if ≥8 patients responded or were progression free at 12 weeks.

Kaplan-Meier estimates of PFS were calculated along with corresponding 95% confidence intervals (CIs). The median PFS and the 12-week PFS rates were determined. Cox regression modeling of PFS was used to asses the effect of perifosine on survival while controlling for other confounders. The objective response rates are reported with their 2-sided exact binomial 95% CIs.

RESULTS

Patient Characteristics

In the Perifosine 228 trial, 24 patients with RCC (median age, 67 years) were enrolled and received treatment with perifosine from April 2007 to February 2009. In the Perifosine 231 patients, 50 patients (32 in Group A and 18 in Group B; median age, 64 years) were enrolled and received treatment with perifosine from December 2007 to July 2011. Accrual to both studies was halted in December 2008 after a decision by the sponsor because of lack of resources. Patient characteristics from both studies are listed in Table 1. In both Perifosine 228 and Perifosine 231, the majority of patients had an ECOG performance status of 1 and clear cell histology (96% and 88%, respectively). In Perifosine 228, patients were distributed equally with respect to the previous receipt of VEGF-targeted TKIs: 12 patients (50%) had received sorafenib, and 12 patients (50%) had received sunitinib. In Perifosine 231, the majority of patients in both groups (76%) had previously received sunitinib; and, in Group B, the majority of patients (67%) had previously received temsirolimus.

Table 1. Patient Characteristics
 No. of Patients (%)
Baseline CharacteristicPerifosine 228Perifosine 231
  1. Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; mTOR, mammalian target of rapamycin; NA, not applicable.

No. of patients24Group A, 32; Group B, 18
Median age [range], y67 [47-78]64 [46-80]
Men16 (67)39 (78)
ECOG PS, n (%)  
 015 (63)20 (40)
 19 (37)30 (60)
Histology  
 Clear cell23 (96)43 (86)
 Nonclear cell1 (4)5 (10)
 Unknown 2 (4)
No. of prior systemic therapies
 112 (50)Group A: 21/32 (66)
 2 Group B: 14/18 (78)
 ≥212 (50)Group A: 11/32 (34)
 ≥3 Group B: 4/18 (22)
Prior tyrosine kinase inhibitor
 Sorafenib12 (50)12 (24)
 Sunitinib12 (50)38 (76)
Prior mTOR inhibitorNA 
 Temsirolimus 12 (67)
 Everolimus 6 (33)

Efficacy

In Perifosine 228, all 24 patients were evaluable for efficacy and toxicity. Overall, the patients received study drug for a range of 2 to 84 weeks. One patient experienced a confirmed objective PR according to RECIST (objective response rate, 4%; 95% CI, 0.7%-20%) (Table 2). Another patient experienced an unconfirmed PR on the first CT scan but developed new bone metastases on the confirmatory scan and, thus, was classified with progressive disease (PD). Eleven patients experienced a best response of SD, whereas 12 patients experienced PD. The overall median PFS was 14.2 weeks (95% CI, 7.7-21.6 weeks) (Fig. 1), and 46% of patients were progression free at 12 weeks.

Table 2. Response to Perifosine
 No. of Patients (%)
  Perifosine 231Combined Analysis
ResponsePerifosine 228Part APart BPerifosine 228 + 321Total
  1. Abbreviations: PD, progressive disease; PR, partial response; SD, stable disease.

PR1 (4)4 (13)1 (6)5 (9)6 (8)
SD11 (46)9 (28)7 (39)20 (36)27 (36)
PD12 (50)19 (59)10 (56)31 (55)41 (55)
Figure 1.

Kaplan Meier progression-free survival (PFS) curves illustrate (A) the survival of patients (Pts) who were treated on the Perifosine 228 trial, (B) the survival of patients who were treated on the Perifosine 231 trial, and (C) a pooled analysis of patients who were treated on both trials. CI indicates confidence interval.

In Perifosine 231, all 50 patients were evaluable for toxicity and efficacy analysis. Overall, the patients received study drug for a range of 1 to 180 weeks. In Group A, 4 patients experienced a confirmed objective PR (objective response rate, 13%). Duration of responses ranged from 13 weeks to 142 weeks. Nine patients experienced a best response of SD, whereas 19 patients experienced PD. The median PFS for Group A was 14.1 weeks (95% CI, 12.6-21.9 weeks), and 41% of patients were progression free at 12 weeks. In Group B, 1 patient experienced a confirmed objective PR (objective response rate, 6%) with a duration of 23 months. Seven patients experienced a best response of SD, whereas 10 patients experienced PD. The median PFS for Group B was 14 weeks (95% CI, 9.4-27.1 weeks), and 44% of patients were progression free at 12 weeks. The overall median PFS for both Group A and Group B was 14 weeks (95% CI, 12.9-20.7 weeks). The objective response rate for the Perifosine 231 trial was 10% (95% CI, 4.5%-22.2%).

The median PFS for all patients who were treated on Perifosine 228 and 231 combined was 14 weeks (95% CI, 12.6-20.0 weeks). The median PFS for patients in Group A of Perifosine 231 combined with all patients in Perifosine 228 was 14.1 weeks (95% CI, 12.6-20.0 weeks). In both studies, for all patients who had previously received sorafenib (n = 24), the median PFS was 19.4 weeks (95% CI, 11.9-30.0 weeks); for all patients who had previously received sunitinib (n = 50), the median PFS was 12.3 weeks (95% CI, 11.4-16.3 weeks). In patients with nonclear cell histology (n = 6), the median PFS was 13.4 weeks (95% CI, 7-19.7 weeks). The objective response rate for all patients who were treated on Perifosine 228 and 231 was 8.1% (95% CI, 3.9%-17%). The median duration for all PRs was 36 weeks.

Toxicity

In both trials, the most common adverse events were gastrointestinal symptoms (diarrhea, nausea), fatigue, and musculoskeletal pain (Table 3). There were very few grade 3 or 4 adverse events in either study, and the most common events were anemia (n = 3), edema (n = 3), and hypomagnesemia (n = 3). There was no clear correlation between adverse events and response to perifosine. Overall, 4 patients (17%) in Perifosine 228 and 10 patients (20%) in Perifosine 231 required a dose reduction to a 50 mg daily because of adverse events.

Table 3. Toxicities
 No. of Patients (%)
 Perifosine 228Perifosine 231
ToxicityGrade 1-2Grade 3-4Grade 1-2Grade 3-4
  1. Abbreviations: GI, gastrointestinal.

Nausea12 (50)1 (4)28 (56)1 (2)
Diarrhea10 (42)0 (0)16 (32)0 (0)
Musculoskeletal pain11 (46)0 (0)10 (20)0 (0)
Fatigue10 (42)0 (0)20 (40)2 (4)
Anorexia5 (21)0 (0)7 (14)0 (0)
Edema5 (21)0 (0)7 (14)1 (2)
Dyspnea5 (21)1 (4)8 (16)2 (4)
Cough5 (21)0 (0)8 (1)0 (0)
Gout4 (17)0 (0)3 (6)0 (0)
Heartburn3 (13)0 (0)0 (0)0 (0)
Hyperglycemia4 (17)0 (0)1 (2)2 (4)
Flatulence3 (13)0 (0)3 (6)0 (0)
Back pain0 (0)1 (4)7 (14)0 (0)
Anemia0 (0)0 (0)7 (14)2 (4)
Pulmonary embolism0 (0)1 (4)0 (0)2 (4)
GI bleed0 (0)1 (40)0 (0)1 (2)
Dehydration0 (0)1 (4)3 (6)0 (0)
Hyponatremia0 (0)1 (4)0 (0)0 (0)
Syncope0 (0)1 (4)0 (0)0 (0)
Constipation0 (0)0 (0)7 (14)0 (0)
Insomnia0 (0)0 (0)6 (12)0 (0)
Increased creatinine0 (0)0 (0)5 (10)0 (0)
Hypomagnesemia0 (0)0 (0)0 (0)3 (6)
Asthenia0 (0)0 (0)7 (14)0 (0)
Hyperuricemia0 (0)0 (0)2 (4)2 (4)
Dizziness0 (0)0 (0)7 (14)0 (0)

DISCUSSION

Despite the many recent advances in RCC therapeutics, currently available agents in RCC are generally limited to 2 classes of drugs: VEGF signaling inhibitors and allosteric inhibitors of mTOR. Immune-based therapy with interferon and interleukin-2 and chemotherapy with fluorinated pyrimidines are not commonly used.10 Novel agents with alternate mechanisms of action are needed to improve the therapeutic outcomes for patients with advanced RCC. In our 2 current phase 2 trials, we describe the efficacy of a novel alkylphospholipid Akt inhibitor in previously treated patients with advanced RCC.

Although neither trial was completely accrued, perifosine exceeded the proscribed efficacy goals in both trials. Both objective tumor responses and prolonged disease stability were observed (Fig. 2). In some patients, responses were quite prolonged, lasting as long as 180 weeks (median PR duration, 36 weeks). At the same time, perifosine also was very well tolerated, as evidenced by the low incidence of grade 3 and 4 toxicities (Table 3). It must be noted, however, that the aforementioned efficacy goals and outcome measures were established shortly after the approval of both sorafenib and sunitinib and in the absence of data on the efficacy of second-line therapy after failure on either of those agents. These data must be put into clinical context with the recent description in randomized phase 3 trials of the activity of everolimus, sorafenib, and axitinib in patients with advanced RCC who had failed on at least 1 or more VEGF-targeted TKI.

Figure 2.

These images are representative of response in lung metastases from a patient who received treatment with perifosine.

In the Renal Cell Cancer Treatment with Oral RAD001 Given Daily (RECORD-1) trial, 416 patients with advanced RCC who had failed prior treatment with sorafenib, sunitinib, or both were randomized to receive either everolimus (n = 277) or placebo (n = 139) with a primary endpoint of PFS.2 At the final assessment, the median PFS for the patients who received everolimus was 4.9 months compared with 1.9 months in the placebo group (hazard ratio, 0.33; 95% CI, 0.25-0.43; P < .0001).11 In the patients who received everolimus, those who had previously received sorafenib alone had a median PFS of 5.9 months versus 3.9 months in those who had previously received sunitinib alone. Five patients (2%) in the everolimus group experienced PRs versus 0 in the placebo group. In the AXIS trial (a trial of the comparative effectiveness of axitinib vs sorafenib in advanced RCC), 723 patients with advanced RCC who had failed 1 prior systemic therapy (sunitinib, bevacizumab, temsirolimus, or cytokine-based therapy) were randomized in a 1:1 fashion to receive either axitinib (n = 361) or sorafenib (n = 362) with a primary endpoint of PFS.12 In patients who had previously received sunitinib, the PFS was 4.8 months in the axitinib group versus 3.8 months in the sorafenib group (P = .0107).

Although comparisons across independent trials are challenging, it must be noted that the patient populations studied in the 2 trials described here (Table 1) were quite similar to one another and also were similar to the population studied in the RECORD-1 trial with respect to basic demographics (age, sex), performance status, and prior therapies (proportion of patients that received prior sunitinib vs sorafenib).2, 11 Although the demographics, performance status, and number of prior therapies among patients in the Perifosine 228 and Perifosine 231 trials also were very similar to those variables among patients in the AXIS study, the comparison is complicated further by the exclusion of prior sorafenib treatment in the AXIS trial.12 However, acknowledging the limitations of historic comparisons, within this more modern clinical context, the efficacy of perifosine that we have described after failure of sorafenib or sunitinib appears modest and not superior to currently available second-line agents.

The exact mechanism for the observed antitumor effect of perifosine in RCC remains unknown. Multiple mechanisms have been implicated that may mediate the antitumor effect of perifosine in preclinical studies, including the induction of apoptosis through both intrinsic and extrinsic pathways, cell cycle arrest, and antiangiogenesis in a variety of tumor types,13-19 all of which are mediated through Akt rather than mTOR. The finding that PFS for patients in Groups A and B on the Perifosine 231 trial was nearly identical supports the belief that the mechanism of action of perifosine in RCC is distinct from that of the allosteric mTOR inhibitors. Furthermore, many investigators have demonstrated additive benefits from combining perifosine with a rapalogue in preclinical studies of multiple tumor types, including RCC.20-22 These findings, combined with the observation that perifosine is very well tolerated with toxicities that generally do not overlap with those of VEGF antagonists and rapalogues, suggest that perifosine may be an attractive agent to assess clinically in combination with currently available agents in RCC. Indeed, a clinical trial is already underway combining perifosine with temsirolimus in patients with malignant glioma (National Clinical Trial identifier NCT01051557).

In conclusion, although perifosine has clear single-agent activity in patients with advanced RCC after failure on VEGF-targeted TKIs, this activity must be considered modest and clearly is not superior to that of other currently available second-line agents. Overall, these results to do not support the further clinical development of perifosine as a single-agent in RCC. Nonetheless, a small subset of patients clearly derived substantial clinical benefit from treatment with perifosine. Further studies are needed into the mechanism of Akt activation in RCC, and such investigations may lead to the better selection of patients who might respond to similarly targeted agents. As perifosine has currently demonstrated promising activity in combination with chemotherapy for gastrointestinal malignancies, combinational therapy with agents like mTOR inhibitors may be worthy of further investigation in RCC.

FUNDING SOURCES

This study was supported by Keryx Pharmaceuticals and by a grant from the National Institute of Health/National Cancer Institute (P50CA101942).

CONFLICT OF INTEREST DISCLOSURES

Peter Sportelli is an employee of Keryx Pharmaceuticals.

Ancillary