Abstract

Background

This phase II study assessed the safety and efficacy of everolimus, an oral mammalian target of rapamycin inhibitor in advanced transitional carcinoma cell (TCC) after failure of platinum-based therapy.

Patients and methods

Thirty-seven patients with advanced TCC received everolimus 10 mg/day until progressive disease (PD) or unacceptable toxicity. The primary end point was the disease control rate (DCR), defined as either stable disease (SD), partial response (PR), or complete response at 8 weeks. Angiogenesis-related proteins were detected in plasma and changes during everolimus treatment were analyzed. PTEN expression and PIK3CA mutations were correlated to disease control.

Results

Two confirmed PR and eight SD were observed, resulting in a DCR of 27% at 8 weeks. Everolimus was well tolerated. Compared with patients with noncontrolled disease, we observed in patients with controlled disease a significant higher baseline level of angiopoietin-1 and a significant early plasma decrease in angiopoietin-1, endoglin, and platelet-derived growth factor-AB. PTEN loss was observed only in patients with PD.

Conclusions

Everolimus showed clinical activity in advanced TCC. The profile of the plasma angiogenesis-related proteins suggested a role of the everolimus antiangiogenic properties in disease control. PTEN loss might be associated with everolimus resistance.

introduction

Bladder cancer is the fourth most common malignancy in men. It is estimated that 69 250 new cases of bladder cancer will be diagnosed in the United States in 2011 with 14,990 cancer-related deaths [1]. Transitional cell carcinoma (TCC) represents the most frequent histological type of bladder cancer, accounting for ∼90% of cases. Locally advanced or metastatic TCC has a poor prognosis with a 5-year overall survival (OS) rate <20% [2]. Platinum-based chemotherapy increases OS in the first-line treatment of metastatic TCC, with a median OS reaching 13–15 months compared with 3–6 months if these patients remain untreated [3, 4]. After platinum failure, only one drug, vinflunine, has been shown to date to improve the median progression free survival (PFS) over best supportive care (BSC) (3.0 versus 1.5 months, P = 0.001) although no statistically significant OS increase in the intent-to-treat population was observed (6.9 versus 4.6 months, P = 0.287) [5]. Due to the lack of therapeutic options, new agents are urgently needed.

Activation of the phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) pathway is detected in a wide range of cancers, including TCC, and results in uncontrolled cell proliferation [6, 7]. The principal causes of PI3K/Akt/mTOR cascade activation are decreased expression of the tumor suppressor phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and deregulation of PI3K/AKT signaling through overexpression or activation of growth factor receptors and activating mutations in PI3K. Inactivation of PTEN is detected in up to 30% of TCC and is associated with tumor aggressiveness and worsened patient outcomes [89]. Activating mutations of PI3K are identified in up to 27% of TCC and occur principally in three ‘hot spots’ of the PIK3CA gene, E545K (52%) and E542K (24%) in the exon 9 and H1047R (13%) in the exon 20 [10–12].

Angiogenesis is implicated in TCC development and depends mainly on the PI3K/Akt/mTOR pathway, via hypoxia-inducible factor and vascular endothelial growth factor (VEGF) production. Elevated VEGF messenger RNA expression in TCC is associated with high stage and grade, vascular invasion, metastases, and poor PFS [13, 14].

Everolimus is a rapamycin derivative, inhibiting mTOR activity, and approved for the treatment of metastatic renal cell carcinoma (RCC), advanced pancreatic neuroendocrine tumors, and inoperable subependymal giant cell astrocytomas associated with tuberous sclerosis [15–17]. Preclinical studies demonstrated that everolimus controls the bladder cancer cell growth in vitro and in vivo through inhibition of the mTOR pathway and antiangiogenic effects [18–20].

We conducted a phase II study of everolimus in patients with locally advanced or metastatic TCC whose disease progressed after platinum-based therapy. The primary end point was the disease control rate (DCR) at 8 weeks after treatment initiation.

patients and methods

eligibility

Eligible patients were required to have histologically or cytologically confirmed locally advanced or metastatic TCC, documented progression after first-line platinum-based chemotherapy (in neoadjuvant/adjuvant setting or for distant metastases), disease not amenable to curative treatment, at least one measurable lesion according to RECIST, and an Eastern Cooperative Oncology Group performance status of zero to two. An interval of at least 4 weeks since the last cytotoxic chemotherapy, biological therapy, surgery, or radiotherapy was required. Patients were required to have an absolute neutrophil count >1500/μl, hemoglobin >9 g/dl, platelet count >100 000/μl, serum creatinine <1.5 × the upper limit of normal (ULN), total bilirubin <1.5 ULN, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) <2.5 ULN, fasting cholesterol <300 mg/dl, and fasting triglycerides <2.5 ULN.

Patients were excluded if they received more than two systemic treatments for their metastatic disease and previous treatment with mTOR inhibitors. Patients were also ineligible if they had significant cardiovascular disease, active infection, known central nervous system metastases, autoimmune disease, or other malignancies within 5 years, excluding adequately treated carcinoma in situ of the cervix or basal or squamous cell skin cancer.

The translational and clinical parts of the study were approved by an independent ethics committee and the Belgian health authority in accordance with European regulations and conducted in accordance with the Declaration of Helsinki (October 2000). Translational research was prospectively planned, and patients provided informed consent for plasma and tumor storage.

study objectives and outcome

This was an open-label, multicenter single-arm phase II study. Patients received everolimus monotherapy 10 mg/day continuously until progressive disease (PD) or unacceptable toxicity.

The primary end point was the DCR, defined as complete response (CR), partial response (PR), or stable disease (SD) at 8 weeks after treatment initiation, according to RECIST (version 1.1). The patients with controlled disease were defined as patients with either PR or SD, and patients with noncontrolled disease as patients with PD at week 8. Secondary end points were toxicity, PFS, and OS. PFS was defined as the time interval between the date of inclusion and the date of PD or death. OS was defined as the time interval between the date of inclusion until death or date of last follow-up.

Baseline evaluation was carried out within 2 weeks before inclusion and included medical history, physical examination, electrocardiogram, and chest and abdominal computed tomography. Laboratory tests included hemogram, creatinine, total protein, albumin, bilirubin, AST and ALT, lactate dehydrogenase, alkaline phosphatase, serum-corrected calcium, fasting glucose, triglycerides, and cholesterol. Imaging was repeated every 8 weeks and centrally reviewed. Adverse events were monitored throughout the study and recorded according to the National Cancer Institute—Common Toxicity Criteria (version 3). The relative dose intensity of everolimus was calculated as the actual dose intensity divided by the planned dose intensity, multiplied by 100.

plasmatic angiogenesis proteins

Hypothesizing that everolimus could have antiangiogenic properties, we carried out angiogenesis-related analyses on plasma samples at baseline (before the treatment) and early during treatment (at day 28), using a Proteome Profiler Array (Angiogenesis Array Kit; R&D Systems, Minneapolis, MN). Each membrane array was incubated with patient plasma and an antibody detection cocktail overnight at 4°C and then with streptavidin-horseradish peroxidase conjugate, followed by a chemiluminescent detection reagent (ECL; Healthcare Life Sciences, Pittsburgh, PA). After exposure to X-ray film, the positive signals were quantified using the ImageQuant TL program (Amersham Bioscience, Piscataway, NJ). For each protein, the results were presented as the percentage of the positive control.

The most significant results were confirmed by ‘sandwich’ enzyme-linked immunosorbent assay (ELISA) following the instructions of the ELISA kits [Angiopoietin-1 Quantikine (DANG10), Endoglin Quantikine (DANG00), and platelet-derived growth factor (PDGF)-AB Quantikine (DHD00B), all from R&D Systems].

PTEN status and PIK3CA mutation

PTEN immunochemistry carried out on archival paraffin-embedded tumor tissues was assessed by a PTEN monoclonal antibody (Cell Signaling, Danvers, MA, #9559). Detection was carried out using a secondary horseradish peroxydase antibody. The tumor cells were assessed for the predominant staining intensity from zero (absent) to three (strong) and for the percentage of stained cells with the predominant staining intensity. For PTEN, positive staining of stromal cells served as an internal control. PTEN-negative tumors showed complete absence of cytoplasmic staining of the malignant cells while maintaining the stromal cell staining. For the PIK3CA mutations detection, DNA was extracted from paraffin-embedded blocks using the DNA extraction kit from Qiagen, Venlo, Netherlands (#51304). Detection of PIK3CA mutations, including E545K, E542K, and H1047R, was carried out by real-time PCR using the PIK3CA Mutation Test Kit from Qiagen (PK-01).

statistical methods

This study employed a Simon optimal one-sample two-stage testing procedure to determine the sample size [21]. Based on hypotheses of Po = 0.10, P1 = 0.25, α = 0.15, and β = 0.10, 37 patients needed to be recruited into the trial. At least 6 of 37 patients were required to achieve CR, PR, or SD at 6–8 weeks after treatment for the study to meet its primary end point. The trial was to be discontinued if ≤1 CR, PR, or SD were observed after 17 patients had been treated for 6–8 weeks. Kaplan–Meier analyses were used to calculate PFS and OS. Student's t-test was used to compare the mean protein plasma levels of the different time points and subgroups. Association between PTEN status and disease control was assessed using the Fisher's exact test.

results

patient characteristics

Between November 2008 and October 2009, 37 patients from 10 Belgian centers were included (Figure 1). All patients had locally advanced or metastatic TCC and were previously exposed to platinum-based chemotherapy in either the neoadjuvant/adjuvant or palliative setting. Baseline characteristics are described in Table 1.

Table 1.

Patients demographic and clinical characteristics

 N = 37 (%) 
Male/Female 28/9 
Age, years  
 Median (range) 63 (32–86) 
ECOG (performance status)  
 0 3 (8%) 
 1 31 (84%) 
 2 3 (8%) 
Tumor grade at diagnosis  
 Well differentiated 3 (8%) 
 Moderately differentiated 7 (19%) 
 Poorly differentiated 20 (54%) 
 Unknown 7 (19%) 
Platinum-based chemotherapy  
 Neoadjuvant 3 (8%) 
 Adjuvant 8 (22%) 
 Palliative 30 (81%) 
 Methotrexate, vinblastine, adriamycine, cisplatine (MVAC regimen) 5 (14%) 
 Platinum/gemcitabin regimen 36 (97%) 
 One previous chemotherapy line for palliation 33 (89%) 
 Two previous chemotherapy lines for palliation 4 (11%) 
 Platinum-free interval inferior to 6 months 17 (46%) 
Location of disease at inclusion  
 Non visceral (nodes and bones) 9 (24%) 
 Visceral 28 (76%) 
  Lung alone 0 (0%) 
  Liver alone 2 (5.5%) 
  Lung and/or liver + nodes + bones 13 (35%) 
  Lung and/or liver + bones 1 (3%) 
  Lung and/or liver + nodes 12 (32.5%) 
Metastatic sites at inclusion  
 Lung 19 (51%) 
 Liver 18 (49%) 
 Nodes 29 (78%) 
 Bones 14 (38%) 
 Other sites (adrenal gland, Retzius space, mesenteric implants, muscle, thyroid) 5 (14%) 
 N = 37 (%) 
Male/Female 28/9 
Age, years  
 Median (range) 63 (32–86) 
ECOG (performance status)  
 0 3 (8%) 
 1 31 (84%) 
 2 3 (8%) 
Tumor grade at diagnosis  
 Well differentiated 3 (8%) 
 Moderately differentiated 7 (19%) 
 Poorly differentiated 20 (54%) 
 Unknown 7 (19%) 
Platinum-based chemotherapy  
 Neoadjuvant 3 (8%) 
 Adjuvant 8 (22%) 
 Palliative 30 (81%) 
 Methotrexate, vinblastine, adriamycine, cisplatine (MVAC regimen) 5 (14%) 
 Platinum/gemcitabin regimen 36 (97%) 
 One previous chemotherapy line for palliation 33 (89%) 
 Two previous chemotherapy lines for palliation 4 (11%) 
 Platinum-free interval inferior to 6 months 17 (46%) 
Location of disease at inclusion  
 Non visceral (nodes and bones) 9 (24%) 
 Visceral 28 (76%) 
  Lung alone 0 (0%) 
  Liver alone 2 (5.5%) 
  Lung and/or liver + nodes + bones 13 (35%) 
  Lung and/or liver + bones 1 (3%) 
  Lung and/or liver + nodes 12 (32.5%) 
Metastatic sites at inclusion  
 Lung 19 (51%) 
 Liver 18 (49%) 
 Nodes 29 (78%) 
 Bones 14 (38%) 
 Other sites (adrenal gland, Retzius space, mesenteric implants, muscle, thyroid) 5 (14%) 
Figure 1.

CONSORT diagram.

Figure 1.

CONSORT diagram.

dose intensity and acute adverse events

The relative dose intensity of everolimus was 78%. The median duration of everolimus treatment was 60 days (range, 17–491 days). Adverse events were consistent with those previously described [15] and are listed in Table 2. The most frequent grade 1–2 toxic effects were anemia (86%), nausea and vomiting (81%), and fatigue (64%). The main grade 3–4 toxic effects were fatigue (27%) and thrombocytopenia (13%). We observed two grade 5 adverse events: a cerebral bleed and a cardiac infarction not related to everolimus according to the investigators. Although both patients were eligible at enrollment according to the inclusion criteria, the cardiac event occurred in a patient with a past medical history of coronary angioplasty and the cerebral event in a patient with cardiovascular risks factors (hypertension, hypercholesterolemia, and coronaropathy) in the absence of thrombocytopenia. For these two patients, everolimus did not induce hypertension or a significant increase in lipid levels.

Table 2.

Everolimus-related adverse events according to National Cancer Institute—Common Toxicity Criteria (version 3)

Toxic effects Overall Grade 1–2 Grade 3–4 Grade 5 
Fatigue 34 (91%) 24 (64%) 10 (27%) 0 (0%) 
Nausea and vomiting 33 (89%) 30 (81%) 3 (8%) 0 (0%) 
Anorexia 12 (32%) 10 (27%) 2 (5%) 0 (0%) 
Stomatitis 14 (38%) 13 (35%) 1 (3%) 0 (0%) 
Diarrhea 15 (40%) 15 (40%) 0 (0%) 0 (0%) 
Anemia 35 (95%) 32 (86%) 3 (8%) 0 (0%) 
Thrombopenia 18 (49%) 13 (35%) 5 (13%) 0 (0%) 
Neutropenia 2 (5%) 2 (5%) 0 (0%) 0 (0%) 
Hyperglycemia 20 (54%) 17 (46%) 3 (8%) 0 (0%) 
Hypertriglyceridemia 19 (51%) 18 (48%) 1 (3%) 0 (0%) 
Liver toxic effects 2 (5%) 1 (3%) 1 (3%) 0 (0%) 
Non infectious pneumonia 2 (5%) 1 (3%) 1 (3%) 0 (0%) 
Cardiac infarction 0 (0%) 0 (0%) 0 (0%) 1 (3%) 
Cerebral bleeding 0 (0%) 0 (0%) 0 (0%) 1 (3%) 
Hypertension 3 (8%) 3 (8%) 0 (0%) 0 (0%) 
Dermatitis 8 (22%) 8 (22%) 0 (0%) 0 (0%) 
Toxic effects Overall Grade 1–2 Grade 3–4 Grade 5 
Fatigue 34 (91%) 24 (64%) 10 (27%) 0 (0%) 
Nausea and vomiting 33 (89%) 30 (81%) 3 (8%) 0 (0%) 
Anorexia 12 (32%) 10 (27%) 2 (5%) 0 (0%) 
Stomatitis 14 (38%) 13 (35%) 1 (3%) 0 (0%) 
Diarrhea 15 (40%) 15 (40%) 0 (0%) 0 (0%) 
Anemia 35 (95%) 32 (86%) 3 (8%) 0 (0%) 
Thrombopenia 18 (49%) 13 (35%) 5 (13%) 0 (0%) 
Neutropenia 2 (5%) 2 (5%) 0 (0%) 0 (0%) 
Hyperglycemia 20 (54%) 17 (46%) 3 (8%) 0 (0%) 
Hypertriglyceridemia 19 (51%) 18 (48%) 1 (3%) 0 (0%) 
Liver toxic effects 2 (5%) 1 (3%) 1 (3%) 0 (0%) 
Non infectious pneumonia 2 (5%) 1 (3%) 1 (3%) 0 (0%) 
Cardiac infarction 0 (0%) 0 (0%) 0 (0%) 1 (3%) 
Cerebral bleeding 0 (0%) 0 (0%) 0 (0%) 1 (3%) 
Hypertension 3 (8%) 3 (8%) 0 (0%) 0 (0%) 
Dermatitis 8 (22%) 8 (22%) 0 (0%) 0 (0%) 

efficacy

Among the 37 treated patients, confirmed PR was observed in 2 patients, SD in 8, and PD in 27, resulting in a DCR of 27% at 8 weeks. Nine patients with rapid clinical PD were unable to be evaluated at 8 weeks and were considered to have PD as prespecified in the protocol. The maximum percentage reduction in the sum of the largest diameters for assessable patients is displayed in Figure 2. Among the eight patients with SD, three experienced some degree of tumor shrinkage (maximum percentage reduction in the sum of the largest diameters of −1%, −2%, and −9%). Disease control was maintained for 4 months in four patients, for 8 months in two patients, and for 14 months in one patient. Interestingly, disease control was also observed in patients with poor prognosis factors, including liver metastases [22] (n = 3), in patients who had not previously responded to chemotherapy (n = 3), and in patients who had a platinum-free interval inferior to 6 months (n = 2). Median PFS was 61 days [95% confidence interval (CI) 49–63 days; range, 17 –491 days], and median OS was 101 days (95% CI 82–128 days; range, 17–571 days) (Figure 3).

Figure 2.

Waterfall plot demonstrating the maximum percentage modification in the sum of the largest diameters of the 28 assessable patients (centrally reviewed). According to RECIST 1.1 criteria, blue plots represent progressive disease, yellow plots represent stable disease, red plots represent partial response, and blue plots * represent stable disease in target lesions but apparition of new metastases.

Figure 2.

Waterfall plot demonstrating the maximum percentage modification in the sum of the largest diameters of the 28 assessable patients (centrally reviewed). According to RECIST 1.1 criteria, blue plots represent progressive disease, yellow plots represent stable disease, red plots represent partial response, and blue plots * represent stable disease in target lesions but apparition of new metastases.

Figure 3.

Progression-free survival and overall survival probabilities.

Figure 3.

Progression-free survival and overall survival probabilities.

angiogenesis-related proteins

In 27 patients out of the 28 radiologically assessable at 8 weeks, we analyzed the angiogenesis-related proteins in baseline and day 28 plasma samples, using a Human Angiogenesis Array Kit. Seventeen proteins were detected. For the whole population, at day 28, we observed a decrease in angiogenic proteins such as PDGF-AB (−67.9%; P < 0.001) and insulin growth factor-binding protein 2 (IGFBP-2) (−13.1%; P = 0.041) and an increase in antiangiogenic proteins such as angiostatin (+113.9%; P < 0.001), pentraxine 3 (+27.8%; P = 0.003), and platelet factor 4 (+40%; P = 0.017). In addition, at day 28, we observed an increase in endoglin level (+32.3%; P = 0.027) compared with baseline (Table 3). Compared with patients with noncontrolled disease (N = 17), patients with controlled disease (N = 10) present a decrease in angiopoietin-1 (−65.2% versus +25.7%; P = 0.016) and endoglin (−9.8% versus +62.8%; P = 0.008) (Table 3). Baseline angiopoietin-1 expression was higher in patients with controlled disease compared with patients with noncontrolled disease (P = 0.040) (Table 4). Of note, both PDGF-AA and PDGF-AB tended to show a higher baseline expression (+34.9% and +56%, respectively; Table 4) and a larger decrease after everolimus (−46.7% versus −9.6% and −80.5% versus −54.8%, respectively; Table 3) in patients with controlled (versus noncontrolled) disease.

Table 3.

Changes in the plasma angiogenesis-related proteins between baseline and day 28

Angiogenesis-related proteins detected with Human Angiogenesis Array Kit Changes in protein expression between baseline and day 28 after treatment. The value represents the variation in % of the mean levels between the baseline and the day 28 plasmatic samples.
 
 Whole population (N = 27), P value, paired Student's t-test Noncontrolled disease, (N = 17) % Controlled disease, (N = 10) % Difference in intrapatient coefficient variation of protein expression between noncontrolled and controlled disease patients. P value, unpaired Student's t-test 
Angiogenin −1.5%, P = 0.610 −2.7 +0.5 3.2%, P = 0.571 
Angiopoietin-1 −36.3%, P = 0.122 +25.7 −65.2 90.9%, P = 0.016 
Angiostatin +113.9%, P < 0.001 +106.9 +128.2 21.3%, P = 0.976 
CXCL16 +0.9%, P = 0.840 +5.9 −7.3 13.2%, P = 0.140 
Endoglin +32.3%, P = 0.027 +62.8 −9.8 72.6%, P = 0.008 
Endostatin +14%, P = 0.075 +21.2 +2.6 18.6%, P = 0.229 
IGFBP-1 +1%, P = 0.788 +2.3 −1.1 3.4%, P = 0.635 
IGFBP-2 −13.1%, P = 0.041 −7 −22.7 15.7%, P = 0.117 
IGFBP-3 −0.1%, P = 0.976 −1.8 +2.8 4.6%, P = 0.333 
MMP-9 +3.5%, P = 0.469 +4.9 +1.1 3.8%, P = 0.698 
Pentraxin-3 +27.8%, P = 0.003 +21.7 +39.5 17.8%, P = 0.452 
PDGF-AA −27.3%, P = 0.099 −9.6 −46.7 37.1%, P = 0.157 
PDGF-AB −67.9%, P < 0.001 −54.8 −80.5 25.7%, P = 0.110 
Platelet factor 4 +40%, P = 0.017 +41.1 +37.6 3.5%, P = 0.742 
Prolactin +32.6%, P = 0.075 +42.4 +12.3 30.1%, P = 0.213 
Serpin-1 −1.6%, P = 0.570 −0.6 −3.0 2.4%, P = 0.649 
Thrombospondin-1 +9.8%, P = 0.652 +13.7 +2.5 11.2%, P = 0.777 
Angiogenesis-related proteins detected with Human Angiogenesis Array Kit Changes in protein expression between baseline and day 28 after treatment. The value represents the variation in % of the mean levels between the baseline and the day 28 plasmatic samples.
 
 Whole population (N = 27), P value, paired Student's t-test Noncontrolled disease, (N = 17) % Controlled disease, (N = 10) % Difference in intrapatient coefficient variation of protein expression between noncontrolled and controlled disease patients. P value, unpaired Student's t-test 
Angiogenin −1.5%, P = 0.610 −2.7 +0.5 3.2%, P = 0.571 
Angiopoietin-1 −36.3%, P = 0.122 +25.7 −65.2 90.9%, P = 0.016 
Angiostatin +113.9%, P < 0.001 +106.9 +128.2 21.3%, P = 0.976 
CXCL16 +0.9%, P = 0.840 +5.9 −7.3 13.2%, P = 0.140 
Endoglin +32.3%, P = 0.027 +62.8 −9.8 72.6%, P = 0.008 
Endostatin +14%, P = 0.075 +21.2 +2.6 18.6%, P = 0.229 
IGFBP-1 +1%, P = 0.788 +2.3 −1.1 3.4%, P = 0.635 
IGFBP-2 −13.1%, P = 0.041 −7 −22.7 15.7%, P = 0.117 
IGFBP-3 −0.1%, P = 0.976 −1.8 +2.8 4.6%, P = 0.333 
MMP-9 +3.5%, P = 0.469 +4.9 +1.1 3.8%, P = 0.698 
Pentraxin-3 +27.8%, P = 0.003 +21.7 +39.5 17.8%, P = 0.452 
PDGF-AA −27.3%, P = 0.099 −9.6 −46.7 37.1%, P = 0.157 
PDGF-AB −67.9%, P < 0.001 −54.8 −80.5 25.7%, P = 0.110 
Platelet factor 4 +40%, P = 0.017 +41.1 +37.6 3.5%, P = 0.742 
Prolactin +32.6%, P = 0.075 +42.4 +12.3 30.1%, P = 0.213 
Serpin-1 −1.6%, P = 0.570 −0.6 −3.0 2.4%, P = 0.649 
Thrombospondin-1 +9.8%, P = 0.652 +13.7 +2.5 11.2%, P = 0.777 

The most relevant proteins are shown in bold type.

IGFBP, insulin growth factor-binding protein; MMP-9, matrix metalloproteinase 9; PDGF, platelet-derived growth factor.

Table 4.

Differences in baseline plasma angiogenesis-related proteins (mean levels) between patients with controlled disease at week 8 and patients with noncontrolled disease

Angiogenesis-related proteins detected with Human Angiogenesis Array Kit Baseline mean levels differences in controlled disease (N = 10) compared with noncontrolled disease (N = 17) patients. P value, unpaired Student's t-test 
Angiogenin +2.5%, P = 0.506 
Angiopoietin-1 +242.5%, P = 0.040 
Angiostatin −15.2%, P = 0.528 
CXCL16 +3.1%, P = 0.586 
Endoglin +25.5%, P = 0.257 
Endostatin +4.0%, P = 0.524 
IGFBP-1 +0.4%, P = 0.778 
IGFBP-2 +8.8%, P = 0.355 
IGFBP-3 +1.0%, P = 0.669 
MMP-9 +0.3%, P = 0.945 
Pentraxin-3 −10.3%, P = 0.541 
PDGF-AA +34.9%, P = 0.348 
PDGF-AB +56.0%, P = 0.196 
Platelet Factor 4 −10.1%, P = 0.514 
Prolactin −17.2%, P = 0.269 
Serpin-1 +5.9%, P = 0.112 
Thrombospondin-1 −0.7%, P = 0.828 
Angiogenesis-related proteins detected with Human Angiogenesis Array Kit Baseline mean levels differences in controlled disease (N = 10) compared with noncontrolled disease (N = 17) patients. P value, unpaired Student's t-test 
Angiogenin +2.5%, P = 0.506 
Angiopoietin-1 +242.5%, P = 0.040 
Angiostatin −15.2%, P = 0.528 
CXCL16 +3.1%, P = 0.586 
Endoglin +25.5%, P = 0.257 
Endostatin +4.0%, P = 0.524 
IGFBP-1 +0.4%, P = 0.778 
IGFBP-2 +8.8%, P = 0.355 
IGFBP-3 +1.0%, P = 0.669 
MMP-9 +0.3%, P = 0.945 
Pentraxin-3 −10.3%, P = 0.541 
PDGF-AA +34.9%, P = 0.348 
PDGF-AB +56.0%, P = 0.196 
Platelet Factor 4 −10.1%, P = 0.514 
Prolactin −17.2%, P = 0.269 
Serpin-1 +5.9%, P = 0.112 
Thrombospondin-1 −0.7%, P = 0.828 

+Represents higher mean value in controlled patients; −represents lesser mean value in controlled patients.

The most relevant proteins are shown in bold type.

IGFBP, insulin growth factor-binding protein; MMP-9, matrix metalloproteinase 9; PDGF, platelet-derived growth factor.

Angiopoietin-1, endoglin as well as PDGF-AB profiles were thus also analyzed with an ELISA method (Figure 4). The patients with controlled disease showed significant decreases in angiopoietin-1 (−80%, P = 0.011), endoglin (−18%, P = 0.010), and PDGF-AB (−54%, P = 0.006) (Figure 4). Moreover, angiopoietin-1 baseline levels were higher in patients with controlled disease than in patients with noncontrolled disease (P = 0.011) (Figure 4A) Furthermore, the differences in intrapatient coefficient of protein expression variation between noncontrolled and controlled disease patients were significant for these three biomarkers: angiopoietin-1 (456 pg/ml; P = 0.013), endoglin (1.59 ng/ml; P = 0.005), and PDGF-AB (890.92 pg/ml; P = 0.017) (supplemental Figure S1, available at Annals of Oncology online).

Figure 4.

Plasma levels (enzyme-linked immunosorbent assay) before and after 28 days of treatment of angiopoietin (A), endoglin (B), and platelet-derived growth factor (PDGF)-AB (C) in noncontrolled disease patients (blue) compared with controlled disease patients (green). The box-whisker plots show the median, interquartile range, 10th, 90th, and 5th and 95th percentiles.

Figure 4.

Plasma levels (enzyme-linked immunosorbent assay) before and after 28 days of treatment of angiopoietin (A), endoglin (B), and platelet-derived growth factor (PDGF)-AB (C) in noncontrolled disease patients (blue) compared with controlled disease patients (green). The box-whisker plots show the median, interquartile range, 10th, 90th, and 5th and 95th percentiles.

PTEN status and PIK3CA mutations

Archival tumor tissues were available for 20 patients, including 6 with controlled disease and 14 with noncontrolled disease at week 8. Absence of PTEN expression was found in 8 of the 14 patients (57%) with noncontrolled disease and in 0% of the patients with controlled disease (supplemental Table S1, available at Annals of Oncology online). PTEN expression was detected in all the 6 patients with controlled disease and in 6 of 14 patients (43%) with noncontrolled disease (supplemental Figure S2, available at Annals of Oncology online). This data suggest an association between PTEN loss and everolimus resistance (P = 0.041). Only three (15%) tumors with a PIK3CA mutation were detected (supplemental Table S1, available at Annals of Oncology online).

discussion

Everolimus was found to have activity in palliative TCC patients. This study met its primary end point as 10 patients (27%) achieved at least SD at 8 weeks. Although median PFS and OS were low (61 days and 101 days, respectively), some patients achieved long-term SD. Even if a randomized phase II study would have eliminated the potential bias of patient selection, this clinical benefit is noteworthy, particularly in this heavily pretreated population for whom there is no available effective treatment. Furthermore, disease control was observed in some patients with poor prognosis factors according to Bellmunt et al. [22] or in some patients with an aggressive and chemoresistant disease. Until recently, there was no established treatment for metastatic TCC after platinum failure. Vinflunine has been approved by the European Medicines Agency for these patients based on the findings that vinflunine improved OS and PFS over BSC, although the OS improvement was statistically significant only in the eligible population and not in the intention-to-treat analysis [5]. Other chemotherapy agents have been investigated in monotherapy in this setting, including paclitaxel, docetaxel, irinotecan, ixabepilone, oxaliplatin, ifosfamide, topotecan, and pemetrexed, with overall response rate (ORR) between 5% and 27.7%, time to progression or PFS ranging from 1.5 to 3.8 months, and median OS ranging from 5 to 9 months [23–30].

Other targeted therapies, including antiangiogenic agents, have also been investigated in this setting with ORR between 0 and 22% [31–34]. Pazopanib, a multitarget tyrosine kinase, appears to be promising based on an interim report, with PR observed in 22%, SD in 61%, and 65% of patients progression-free at 2 months [35]. Another phase II study presented at the American Society of Clinical Oncology (ASCO) Annual Meeting 2011 evaluated everolimus in advanced TCC after failure of platinum-based therapy [36, 37]. Median PFS and OS were 3.3 and 10.5 months. However, 8 of the 43 patients enrolled in this trial who received one or less treatment cycle secondary to rapid progression or toxicity unrelated to everolimus were deemed inevaluable for PFS and were replaced. This probably accounts for the differences in PFS and OS between the trial reported here and the one presented at the ASCO meeting 2011.

Treatment with everolimus was well tolerated in our study population. Thrombocytopenia (13%) was the most frequent grade 3–4 adverse event. This frequency is higher than that reported in the RCC phase III study [15] and may be linked to the bone marrow impairment related to the previous use of chemotherapy, which is not administered in RCC. Two grade 5 adverse events, one cardiac infarction, and one cerebral bleed occurred during treatment. Although atheroembolic and hemorrhagic events are unlikely with mTOR inhibitors [15–17], we cannot exclude that these events could be related to everolimus since our translational research suggested that everolimus has antiangiogenic properties.

Although exploratory and limited by the low number of patients, our translational research suggests new hypotheses on mTOR inhibitor mechanisms.

The plasmatic analyses suggest antiangiogenic properties of everolimus, with decrease in proangiogenic proteins and increase in antiangiogenic proteins (Table 3). Furthermore, the decrease of some angiogenic proteins in patients with controlled disease could reflect the role of angiogenesis inhibition in disease control. Compared with patients with noncontrolled disease, a decrease in angiopoietin-1, PDGF-AB, and endoglin were observed in patients with controlled disease (Figure 4). Angiopoietin-1 and PDGF-AB are two proteins implicated in vessel maturation and stabilization [38–40] and endoglin is a proliferation-associated antigen expressed only by dividing endothelial cells in newly forming tumor vessels [41]. These observations support the hypothesis that everolimus could control cancer growth through an inhibition of vessel proliferation but also of vessel maturation.

Because PTEN acts as a break of Akt activation, PTEN loss is associated with constitutional activation of PI3K/Akt/mTOR signaling. Although some preclinical models demonstrated that PTEN loss could sensitize tumor cells to mTOR inhibitors [42], clinical studies in renal, endometrial, and breast cancers did not show any correlation [43–45]. Furthermore, a recent in vitro study on breast cancer cells showed that cells with PTEN loss were not sensitive to mTOR inhibitors [46]. A clinical trial carried out in PTEN-deficient glioblastoma suggested that rapamycin induced Akt activation, which was associated with shorter PFS [47]. In our study, PTEN loss was found exclusively in patients with noncontrolled disease at week 8. This observation could highlight the importance of the negative feedback mediated by S6 Kinase 1 toward the PI3K/Akt pathway [48]. Indeed, it could be postulated that everolimus induces Akt activation via the removal of this negative feedback. Akt activation could be sustained in case of PTEN loss and therefore might act as a cancer cell survival mechanism inducing everolimus resistance.

There are some limitations in our study interpretation. The number of patients is limited and our trial was not designed and powered to study the role of angiogenesis-related proteins, PTEN status, or PIK3CA mutations as biomarkers. Another limitation is that PTEN status and PIK3CA mutations were carried out predominantly on archival tissue and not on fresh biopsies.

In conclusion, everolimus has clinical activity in advanced TCC after failure of platinum-based chemotherapy. Angiopoietin-1, endoglin, and PDGF-AB are potential biomarkers that should be investigated further in clinical trials. PTEN loss might be associated with everolimus resistance.

funding

Novartis and Televie (FC84766 to ES).

disclosure

Prof. JPM had an advisory role for Genmab and Boehringer and received research fundings from Pfizer and Novartis. The other authors have declared no conflict of interest.

acknowledgements

We thank all the patients, families, nurses, and other collaborators who made this research possible. The authors also like to acknowledge the assistance of Nadia Cappoen and Aline Gillain in the data collection.

references

1
American Cancer Society
Cancer Facts & Figures
 , 
2011
Atlanta, GA
American Cancer Society 2011
2
Bajorin
DF
Dodd
PM
Mazundar
M
, et al.  . 
Long term survival in metastatic transitional cell carcinoma and prognostic factors predicting outcome of therapy
J Clin Oncol
 , 
1999
, vol. 
17
 (pg. 
3173
-
3181
)
3
Saxman
SB
Propert
KJ
Einhorn
LH
, et al.  . 
Long-term follow-up of a phase III intergroup study of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study
J Clin Oncol
 , 
1997
, vol. 
15
 (pg. 
2564
-
2569
)
4
von der Maase
H
Sengelov
L
Roberts
JT
, et al.  . 
Long-term survival results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer
J Clin Oncol
 , 
2005
, vol. 
23
 (pg. 
4602
-
4608
)
5
Bellmunt
J
Theodore
C
Demkov
T
, et al.  . 
Phase III trial of vinflunine plus best supportive care compared with best supportive care alone after a platinum-containing regimen in patients with advanced transitional cell carcinoma of the urothelial tract
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
4454
-
4461
)
6
Garcia
JA
Danielpour
D
Mammalian target of rapamycin inhibition as a therapeutic strategy in the management of urologic malignancies
Mol Cancer Ther
 , 
2008
, vol. 
7
 (pg. 
1347
-
1354
)
7
Chen
M
Cassidy
A
Gu
J
, et al.  . 
Genetic variations in PI3K-AKT-mTOR pathway and bladder cancer risk
Carcinogenesis
 , 
2009
, vol. 
30
 (pg. 
2047
-
2052
)
8
Aveyard
JS
Skilleter
A
Habuchi
T
, et al.  . 
Somatic mutation of PTEN in bladder carcinoma
Br J Cancer
 , 
1999
, vol. 
80
 (pg. 
904
-
908
)
9
Saal
LH
Johansson
P
Holm
K
, et al.  . 
Poor prognosis in carcinoma is associated with a gene expression signature of aberrant PTEN tumor suppressor pathway activity
Proc Natl Acad Sci USA
 , 
2007
, vol. 
104
 (pg. 
7564
-
7569
)
10
Knowles
MA
Platt
FM
Ross
RL
, et al.  . 
Phosphatidylinositol 3-kinase (PI3K) pathway activation in bladder cancer
Cancer Metastasis Rev
 , 
2009
, vol. 
28
 (pg. 
305
-
316
)
11
Platt
FM
Hurst
CD
Tayler
CF
, et al.  . 
Spectrum of phosphatidylinositol 3-kinase pathway gene alterations in bladder cancer
Clin Cancer Res
 , 
2009
, vol. 
15
 (pg. 
6008
-
6017
)
12
Qian
CN
Furge
KA
Knol
J
, et al.  . 
Activation of the PI3K/Akt pathway induces urothelial carcinoma of the renal pelvis: identification in human tumors and confirmation in animal models
Cancer Res
 , 
2009
, vol. 
69
 (pg. 
8256
-
8264
)
13
Wu
W
Shu
X
Hovsepyan
H
, et al.  . 
VEGF receptor expression and signalling in human bladder tumors
Oncogene
 , 
2003
, vol. 
22
 (pg. 
3361
-
3370
)
14
Yang
CC
Chu
KC
Yeh
WM
The expression of vascular endothelial growth factor in transitional cell carcinoma of urinary bladder is correlated with cancer progression
Urol Oncol
 , 
2004
, vol. 
22
 (pg. 
1
-
6
)
15
Motzer
RJ
Escudier
B
Oudard
S
, et al.  . 
Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial
Lancet
 , 
2008
, vol. 
372
 (pg. 
449
-
456
)
16
Yao
JC
Shah
MH
Ito
T
, et al.  . 
Everolimus for advanced pancreatic neuroendocrine tumors
N Engl J Med
 , 
2011
, vol. 
364
 (pg. 
514
-
523
)
17
Krueger
DA
Care
MM
Holland
K
, et al.  . 
Everolimus for subependymal giant-cell astrocytomas in tuberous sclerosis
N Engl J Med
 , 
2010
, vol. 
363
 (pg. 
1801
-
1811
)
18
Fechner
G
Classen
K
Schmidt
D
, et al.  . 
Rapamycin inhibits in vitro growth release of angiogenetic factors in human bladder cancer
Urology
 , 
2009
, vol. 
73
 (pg. 
665
-
668
)
19
Mansure
JJ
Nassim
R
Chevalier
S
, et al.  . 
Inhibition of mammalian target of rapamycin as a therapeutic strategy in the management of bladder cancer
Cancer Biol Ther
 , 
2009
, vol. 
8
 (pg. 
2339
-
2347
)
20
Chiong
E
Lee
IL
Dadbin
A
, et al.  . 
Effects of mTOR inhibitor everolimus (RAD001) on bladder cancer cells
Clin Cancer Res
 , 
2011
, vol. 
17
 (pg. 
2863
-
2873
)
21
Simon
R
Optimal two-stage designs for phase II clinical trials
Control Clin Trials
 , 
1989
, vol. 
10
 (pg. 
1
-
10
)
22
Bellmunt
J
Choueiri
TK
Fougeray
R
, et al.  . 
Prognostic factors in patients with advanced transitional cell carcinoma of the urothelial tract experiencing treatment failure with platinum-containing regimens
J Clin Oncol
 , 
2010
, vol. 
28
 (pg. 
1850
-
1855
)
23
Beer
TM
Goldman
B
Nichols
CR
, et al.  . 
Southwest Oncology Group phase II study of irinotecan in patients with advanced transitional cell carcinoma of the urothelium that progressed after platinum based chemotherapy
Clin Genitourin Cancer
 , 
2008
, vol. 
6
 (pg. 
36
-
39
)
24
Winquist
E
Vokes
E
Moore
MJ
, et al.  . 
A phase II study of oxaliplatin in urothelial cancer
Urol Oncol
 , 
2005
, vol. 
23
 (pg. 
150
-
154
)
25
Witte
RS
Manola
J
Burch
PA
, et al.  . 
Topotecan in previously treated advanced urothelial carcinoma: an ECOG phase II trial
Invest New Drugs
 , 
1998
, vol. 
16
 (pg. 
191
-
195
)
26
Lorusso
V
Pollera
CF
Antimi
M
, et al.  . 
A phase II study of gemcitabine in patients with transitional cell carcinoma of the urinary tract previously treated with platinum. Italian Co-operative Group on Bladder Cancer
Eur J Cancer
 , 
1998
, vol. 
34
 (pg. 
1208
-
1212
)
27
Vaughn
DJ
Broome
CM
Hussain
M
, et al.  . 
Phase II trial of weekly paclitaxel in patients with previously treated advanced urothelial cancer
J Clin Oncol
 , 
2002
, vol. 
20
 (pg. 
937
-
940
)
28
McCaffrey
JA
Hilton
S
Mazumdar
M
, et al.  . 
Phase II trial of docetaxel in patients with advanced or metastatic transitional-cell carcinoma
J Clin Oncol
 , 
1997
, vol. 
15
 (pg. 
1853
-
1857
)
29
Dreicer
R
Li
S
Manola
J
, et al.  . 
Phase 2 trial of epothilone B analog BMS-247550 (ixabepilone) in advanced carcinoma of the urothelium (E3800): a trial of the Eastern Cooperative Oncology Group
Cancer
 , 
2007
, vol. 
110
 (pg. 
759
-
763
)
30
Sweeney
CJ
Roth
BJ
Kabbinavar
FF
, et al.  . 
Phase II study of pemetrexed for second-line treatment of transitional cell cancer of the urothelium
J Clin Oncol
 , 
2006
, vol. 
24
 (pg. 
3451
-
3457
)
31
Wulfing
C
Machiels
JP
Richel
DJ
, et al.  . 
A single-arm, multicenter, open-label phase 2 study of lapatinib as the second-line treatment of patients with locally advanced or metastatic transitional cell carcinoma
Cancer
 , 
2009
, vol. 
115
 (pg. 
2881
-
2890
)
32
Twardowski
P
Stadler
W
Frankel
P
, et al.  . 
Phase II study of aflibercept (VEGF-Trap) in patients (pts) with recurrent or metastatic transitional cell carcinoma (TCC) of the urothelium: A California Cancer Consortium trial
J Clin Oncol
 , 
2009
, vol. 
27
 pg. 
15s
  
(Abstr e16030)
33
Dreicer
R
Li
H
Stein
M
, et al.  . 
Phase 2 trial of sorafenib in patients with advanced urothelial cancer: a trial of the Eastern Cooperative Oncology Group
Cancer
 , 
2009
, vol. 
115
 (pg. 
4090
-
4095
)
34
Gallagher
DJ
Milowsky
MI
Gerst
SR
, et al.  . 
Phase II study of sunitinib in patients with metastatic urothelial cancer
J Clin Oncol
 , 
2010
, vol. 
28
 (pg. 
1373
-
1379
)
35
Necchi
A
Nicolai
N
Guglielmi
A
, et al.  . 
Phase II study of pazopanib monotherapy for patients with relapsed/refractory urothelial cancer.
 
Presented at the 35th European Society for Medical Oncology (ESMO) Congress. Milan, Italy, October 8–12, 2010
36
Milowsky
MI
Regazzi
AM
Garcia-Grossman
IR
, et al.  . 
Final results of a phase II study of everolimus (RAD001) in metastatic transitional cell carcinoma (TCC) of the urothelium
J Clin Oncol
 , 
2011
, vol. 
29
 pg. 
15s
  
(Abstr 4606)
37
Milowsky
MI
Regazzi
AM
Garcia-Grossman
IR
, et al.  . 
Phase II study of everolimus (RAD001) in metastatic transitional cell carcinoma (TCC) of the urothelium
J Clin Oncol
 , 
2011
, vol. 
29
 pg. 
7s
  
(Abstr 245)
38
Satoh
N
Yamada
Y
Kinugasa
Y
, et al.  . 
Angiopoietin-1 alters tumor growth by stabilizing blood vessels or by promoting angiogenesis
Cancer Sci
 , 
2008
, vol. 
99
 (pg. 
2373
-
2379
)
39
Oka
N
Yamamoto
Y
Takahashi
M
, et al.  . 
Expression of angiopoietin-1 and -2, and its clinical significance in human bladder cancer
BJU Int
 , 
2005
, vol. 
95
 (pg. 
660
-
663
)
40
Rosa
AI
Gonçalves
J
Cortes
L
, et al.  . 
The angiogenic factor angiopoietin-1 is a proneurogenic peptide on subventricular zone stem/progenitor cells
J Neurosci
 , 
2010
, vol. 
30
 (pg. 
4573
-
4584
)
41
Takahashi
N
Kawanishi-Tabata
R
Haba
A
, et al.  . 
Association of serum endoglin with metastasis in patients with colorectal, breast, and other solid tumors, and suppressive effect of chemotherapy on the serum endoglin
Clin Cancer Res
 , 
2001
, vol. 
7
 (pg. 
524
-
532
)
42
Neshat
MS
Mellinghoff
IK
Tran
C
, et al.  . 
Enhanced sensitivity of PTEN-deficient tumors to inhibition of FRAP/mTOR
Proc Natl Acad Sci USA
 , 
2001
, vol. 
98
 (pg. 
10314
-
10319
)
43
Cho
D
Signoretti
S
Dabora
S
, et al.  . 
Potential histologic and molecular predictors of response to temsirolimus in patients with advanced renal cell carcinoma
Clin Genitourin Cancer
 , 
2007
, vol. 
5
 (pg. 
379
-
385
)
44
Oza
AM
Elit
L
Tsao
MS
, et al.  . 
Phase II study of temsirolimus in women with recurrent or metastatic endometrial cancer: a trial of the NCIC clinical trial group
J Clin Oncol
 , 
2011
, vol. 
29
 (pg. 
3278
-
3285
)
45
Ellard
SL
Clemons
M
Gelmon
KA
, et al.  . 
Randomized phase II study comparing two schedules of everolimus in patients with recurrent/metastatic breast cancer: NCIC Clinical Trials Group IND.163
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
4536
-
4541
)
46
Weigelt
B
Warne
PH
Downward
J
PIK3CA mutation, but not PTEN loss of function, determines the sensitivity of breast cancer cells to mTOR inhibitory drugs
Oncogene
 , 
2011
, vol. 
30
 (pg. 
3222
-
3233
)
47
Cloughesy
TF
Yoshimoto
K
Ngiemphu
P
, et al.  . 
Antitumor activity of rapamycin in a phase I trial for patients with recurrent PTEN-deficient glioblastoma
PLoS Med
 , 
2008
, vol. 
5
 pg. 
e8
 
48
Wan
X
Harkavy
B
Shen
N
, et al.  . 
Rapamycin induces feedback activation of Akt signaling through an IGF-1R-dependent mechanism
Oncogene
 , 
2007
, vol. 
26
 (pg. 
1932
-
1940
)