- Email: [email protected]
Phase I safety, pharmacokinetic, and pharmacodynamic study of the oral phosphatidylinositol-3-kinase and mTOR inhibitor BGT226 in patients with advanced solid tumors
original articles
Annals of Oncology
Annals of Oncology 23: 2399–2408, 2012 doi:10.1093/annonc/mds011 Published online 22 February 2012
Phase I safety, pharmacokinetic, and pharmacodynamic study of the oral phosphatidylinositol-3-kinase and mTOR inhibitor BGT226 in patients with advanced solid tumors B. Markman1,†, J. Tabernero1, I. Krop2, G. I. Shapiro2, L. Siu3, L. C. Chen4, M. Mita5, M. Melendez Cuero6, S. Stutvoet6, D. Birle7, Ö. Anak6,‡, W. Hackl6 & J. Baselga1, §*
Received 7 June 2011; revised 22 November 2011; accepted 10 January 2012
Background: This phase I dose-escalation study investigated the maximum tolerated dose (MTD), safety, pharmacokinetics, pharmacodynamics (PDs), and preliminary antitumor activity of BGT226, a potent, oral dual phosphatidylinositol-3-kinase (PI3K)/mammalian target of rapamycin inhibitor. Patients and methods: Fifty-seven patients with advanced solid tumors received BGT226 2.5–125 mg/day three times weekly (TIW). Dose escalation was guided by an adaptive Bayesian logistic regression model with overdose control. Assessments included response per RECIST, [18F]-fluorodeoxyglucose uptake, and phosphorylated-S6 in skin and paired tumor samples. Results: Three patients (125 mg cohort) had dose-limiting toxic effects (grade 3 nausea/vomiting, diarrhea). BGT226related adverse events included nausea (68%), diarrhea (61%), vomiting (49%), and fatigue (19%). BGT226 demonstrated rapid absorption, variable systemic exposure, and a median half-life of 6–9 h. Seventeen patients (30%) had stable disease (SD) as best response. Nine patients had SD for ≥16 weeks. Thirty patients (53%) achieved stable metabolic disease as assessed by [18F]-fluorodeoxyglucose–positron emission tomography; however, no correlation between metabolic response and tumor shrinkage according to computed tomography was observed. PD changes suggested PI3K pathway inhibition but were inconsistent. Conclusions: The MTD of BGT226 was 125 mg/day TIW, and the clinically recommended dose was 100 mg/day TIW. Limited preliminary antitumor activity and inconsistent target inhibition were observed, potentially due to low systemic exposure. Key words: anticancer agent, BGT226, dual inhibitor, mTOR catalytic inhibitor, PI3K pathway, solid tumors
introduction The phosphatidylinositol-3-kinase (PI3K) signal transduction pathway regulates key physiologic functions, including cellular proliferation, growth, metabolism, and survival [1, 2]. It is *Correspondence to: Dr J. Baselga, Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Massachusetts General Hospital, 55 Fruit Street, Lawrence House 108, Boston, MA 02114, USA. Tel: +1-617-643-2438; Fax: +1-617-643-2683; E-mail: [email protected] †
Present address: Monash Institute of Medical Research, Monash University, Melbourne, Australia.
‡
Present address: Oncology Global Development, Novartis Pharma AG, Basel, Switzerland.
§
Present address: Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Boston, USA.
frequently activated in cancer cells, via several mechanisms, including mutation of PI3K signaling components and downstream effectors [3, 4]. Activating mutations in PIK3CA, which encodes the p110α catalytic subunit of PI3K, are present in many human cancers, including breast, colon, endometrial, and hepatocellular cancers [3–5]. Evidence suggests that PI3K pathway activation is oncogenic and may contribute to therapeutic resistance to chemotherapy and targeted agents in a number of cancers, including breast and lung cancer [3, 6–9]. These data suggest that inhibition of PI3K signaling is an attractive target for anticancer therapy, and preclinical studies of PI3K signaling inhibitors demonstrated antiproliferative and antitumor effects [3, 10–12]. BGT226 (Novartis Pharma AG, Basel, Switzerland), an imidazoquinoline derivative, is an orally bioavailable potent
© The Author 2012. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
1 Medical Oncology Department, Vall d’Hebron University Hospital, Universitat Autònoma de Barcelona, Barcelona, Spain; 2Department of Medical Oncology, DanaFarber Cancer Institute, Boston, USA; 3Department of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Canada; 4Nevada Cancer Institute, Las Vegas; 5Cancer Therapy and Research Center, The University of Texas Health Science Center at San Antonio, San Antonio, USA; 6Oncology Translational Medicine, Novartis Pharma AG, Basel, Switzerland; 7Novartis Institutes for Biomedical Research, Cambridge, USA
original articles
patients and methods See supplemental Appendix (available at Annals of Oncology online) for additional details.
The starting dose of 2.5 mg PO on a ‘q48h’ schedule is based on in vivo experiments in rat and dog, which showed a long-lasting effect of BGT226 and a similar efficacy to a q24h schedule (see supplemental Appendix, available at Annals of Oncology online, for additional details). Dose escalation was guided by an adaptive Bayesian logistic regression model with overdose control and proceeded by enrolling cohorts of three or more patients [13, 14]. The dose-limiting toxicity (DLT) rate was evaluated after the first treatment cycle. The dose of the subsequent cohort was chosen to fulfill the overdose criterion of a < 25% chance of the DLT rate exceeding 33%. Dose escalation was further based on review of clinical safety, PK, and laboratory data. DLT was defined as a suspected drug-related toxicity according to National Cancer Institute—Common Terminology Criteria for Adverse Events (NCI–CTCAE) V3.0, unless otherwise specified, occurring < 28 days following the first BGT226 dose. The MTD was defined as the highest dose resulting in DLTs in ≤ 33% of patients in cycle 1.
safety and response assessments Safety assessments were conducted at baseline and weekly during the first eight treatment weeks and every 2 weeks thereafter. Adverse events (AEs) were graded using NCI–CTCAE, unless otherwise specified (supplemental Table S1, available at Annals of Oncology online). All patients had medical history, physical examination, vital signs, hematology and chemistry profile, cardiac enzymes, and electrocardiogram (ECG) assessments. Fasting serum insulin, serum C-peptide, and fasting plasma glucose (FPG) were analyzed weekly during cycle 1 and on days 15 and 28 in subsequent cycles. Hemoglobin A1c and fructosamine were measured at baseline and day 1 of subsequent cycles, starting with cycle 2. Urine glucose was monitored daily in cycle 1. Response was assessed according to the RECIST V1.0 [15] every 8 weeks by investigator and central review. The metabolic antitumor activity of BGT226 was assessed by [18F]fluorodeoxyglucose (FDG)–positron emission tomography (PET), evaluating change from baseline in the sum of maximum standardized uptake values (sSUVmax) in sequential PET scans carried out at baseline and day 28 of cycle 1.
study design The primary objective of this multicentre, open-label phase I doseescalation study was to determine the MTD of oral single-agent BGT226 administered to adult patients with solid tumors. Secondary objectives included evaluation of safety and tolerability, measurement of PK end points, assessment of antitumor activity, and PD effects. Main inclusion criteria were histologically confirmed advanced solid tumors unresponsive to standard therapy, age ≥18 years, life expectancy ≥12 weeks, World Health Organization performance status two or less, and adequate bone marrow, hepatic, and renal function. Key exclusion criteria included prior treatment with PI3K inhibitors, diabetes mellitus, brain metastases, gastrointestinal disease that could affect BGT226 absorption, progressive eye disease, thyroid disease requiring treatment, acute or chronic liver or renal disease, or pancreatitis. The study was conducted in accordance with the International Conference on Harmonisation Good Clinical Practice Guidelines and the Declaration of Helsinki, with approval by ethics committees and health authorities of participating institutions. All patients provided written informed consent.
treatment and dose-escalation criteria Patients were instructed to take BGT226 (hard gelatin capsules) as a single oral daily dose, three times weekly (TIW) (every other day), 2 h after a light breakfast, and to continue fasting for two more hours. Treatment was continued on 28-day treatment cycles until unacceptable toxicity, disease progression, or consent withdrawal.
| Markman et al.
PK analysis Plasma BGT226 levels were measured on days 1 and 8 of cycle 1, day 1 of cycle 2, predose, and at indicated time points post dose using a validated liquid chromatography–tandem mass spectrometry assay with a lower quantification limit of 10 pg/ml. PK parameters were calculated by noncompartmental analysis using WinNonlin® (Pharsight, Mountain View, CA).
PD assessments Immunohistochemical analysis of phosphorylated Akt (serine 473; pAkt), phosphorylated ribosomal protein S6 (serine 240/244; pS6), and Ki-67 were carried out in available pre- and posttreatment skin biopsies and optional paired fresh pre- and posttreatment tumor biopsies. Blood samples were analyzed for levels of circulating angiogenic markers and markers of cell death.
results Fifty-seven patients across five sites were treated with BGT226 between December 2007 and February 2010 (Table 1). Breast (23%), prostate (14%), and colon cancers (12%) were the most common malignancies. All patients were evaluated for safety, response, and PK.
dose escalation and safety findings Patients were enrolled into sequential cohorts at BGT226 dose levels of 2.5, 5, 10, 20, 40, 80, and 125 mg. In the 125-mg
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
pan-class I PI3K and mammalian target of rapamycin (mTOR) catalytic inhibitor. BGT226 is among the first in a novel class of agents targeting the PI3K pathway. BGT226 specifically inhibits p110α, β, δ, and γ, with a preference for the α-isoform (wild type and mutated), and mTOR (Novartis, data on file), with no significant inhibitory activity against other tested kinases (see supplemental Appendix, available at Annals of Oncology online). The IC50 of BGT226 for the individual isoforms of the class I PI3Ks is (±standard deviation) PI3Kα 4 nM (±1), PI3Kβ 63 nM (±10), and PI3Kγ 38 nM (±23) as determined by filter-binding assay (Novartis, data on file). In preclinical studies, BGT226 inhibited cancer cell proliferation with IC50 values in the low nanomolar range and inhibited solid tumor growth in various mouse xenograft models, including glioblastoma multiforme, breast, and prostate cancer (Novartis, data on file). In vivo pharmacokinetic (PK)/ pharmacodynamic (PD) analyses showed a good correlation between BGT226 dose, blood, and tissue levels, effect on the PI3K pathway (determined by Akt phosphorylation), and antitumor activity (Novartis, data on file). Here, we report the results of a phase I, first-in-human doseescalation study of BGT226 to determine the maximum tolerated dose (MTD) when administered orally (PO) to adults with advanced solid tumors. Safety, tolerability, PK/PD properties, and preliminary clinical activity of BGT226 were also evaluated.
Annals of Oncology
original articles
Annals of Oncology Table 1. Baseline patient characteristics Characteristics
60.0 23–82 24 (42.1) 33 (57.9) 28 (49.1) 26 (45.6) 3 (5.3) 13 (22.8) 8 (14.0) 7 (12.3) 4 (7.0) 4 (7.0) 3 (5.3) 2 (3.5) 2 (3.5) 2 (3.5) 12 (21.1) 1.4 1–8 1 (1.8) 56 (98.2) 57 (100) 4 1–13 32 (56.1)
a
One patient each had bladder, cervix, cholangiocarcinoma, kidney, fallopian tubes, mucosal melanoma, pancreas, mesothelioma, rectum, salivary gland, and stomach cancer, and gastrointestinal stromal tumor. ECOG, Eastern Cooperative Oncology Group; WHO, World Health Organization.
cohort, 3 of 11 patients had DLTs, all gastrointestinal in nature: grade 3 diarrhea, grade 3 vomiting, and grade 3 diarrhea and vomiting (Table 2). Based on the DLT incidence, the MTD of BGT226 was defined at 125 mg/day, TIW. As a result of the gastrointestinal AEs associated with the 125 mg dose, a deescalated 100 mg dose level was explored, which resulted in improved tolerance. BGT226 100 mg/day, TIW, was therefore declared the clinically recommended dose. Overall, patients were exposed to BGT226 for a median of 7.6 weeks (Table 3). As of 10 February 2010, all patients had discontinued treatment. Four patients (7%) discontinued due to AEs; two of these were considered treatment related (diarrhea with vomiting, one patient and diarrhea with nausea/ vomiting, one patient, both in the 125-mg cohort), and two were assessed as nontreatment related (sepsis of urological origin, one patient, 125-mg cohort and pathological fracture, one patient, 80-mg cohort). Twelve patients (21%) required dose adjustment or interruption due to AEs.
Volume 23 | No. 9 | September 2012
pharmacokinetics BGT226 was rapidly absorbed after single-dose oral administration, with median peak plasma concentrations (Cmax) observed between 1.0 and 4.6 h post dose (Tmax) (Table 5). Interpatient variability in Cmax and area under the plasma concentration time curve (AUC0 − 48) was relatively doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Age, years Median Range Sex, n (%) Male Female WHO/ECOG performance status, n (%) 0 1 2 Primary tumor, n (%) Breast Prostate Colon Soft tissue sarcoma Uterus Bone Ovary Esophagus Skin melanoma Othera Time since most recent recurrence/relapse, months Median Range Current stage, n (%) Stage III Stage IV Prior antineoplastic therapy, n (%) Number of regimens Median Range Patients with more than three prior regimens, n (%)
No. of patients (N = 57)
Gastrointestinal disorders were the most frequently reported AEs regardless of study drug relationship and included nausea, diarrhea, vomiting, decreased appetite, and abdominal pain (Table 4). Other common AEs included fatigue (33%) and anemia (25%). Other hematological abnormalities were less common (lymphopenia, 10.5%; thrombocytopenia, 3.5%; and neutropenia, 1.8%). Fifty patients (88%) experienced suspected treatment-related AEs. The most common were nausea (68%), diarrhea (61%), vomiting (49%), fatigue (19%), and anorexia (16%) (Table 4). Eleven patients (19%) experienced grade 3/4 drug-related AEs, of which diarrhea was the most common (12%). The other frequent, suspected drug-related AEs were generally mild to moderate in severity. The incidence of drug-related vomiting, nausea, and diarrhea was dose dependent and generally limited to dosing days. Only after the first onset of vomiting, nausea, and/or diarrhea, patients were pretreated and/or treated with antiemetics and/or antidiarrheals, respectively, according to the site’s standards. Medications used included, among others, loperamide and lomotil for diarrhea and aprepitant, metoclopramide, palonosetron, lorazepam, and prochlorperazine for vomiting and nausea. Despite pretreatment and/or treatment of these events, the 125 mg dose was not well tolerated. Generally, the 100 mg dose was well tolerated due to these measures. The only hematological toxicity potentially related to BGT226 was anemia (11%). Grade 3 anemia regardless of relationship was observed in four patients (two in the 2.5-mg cohort, one each in the 10-mg and 100-mg cohorts). For these patients, grade 1 or 2 anemia was already noted at screening and/or baseline, and hemoglobin was low throughout the study. Most changes in biochemical parameters were mild in severity (grade 1). The most frequent grade 2 or higher biochemical abnormalities were aspartate aminotransferase (grade 2, N = 9; grade 3, N = 3) and alkaline phosphatase (grade 2, N = 4; grade 3, N = 4) elevations. Two patients had newly occurring or grade 4 hyperuricemia. No other grade 4 abnormalities of biochemistry parameters were noted. Two patients had grade 1 new or worsened FPG (10-mg and 100mg cohorts). No grade > 1 FPG abnormality was observed in any patient. No clinically significant abnormalities were observed by cardiac imaging or ECG after treatment initiation. Serious AEs (SAEs) were experienced by 10 patients (17.5%) (supplemental Table S2, available at Annals of Oncology online). One patient (125-mg cohort) experienced SAEs considered by investigators to be treatment related (diarrhea, nausea, and vomiting). There were no deaths on BGT226 treatment. Four patients died within 28 days after the last BGT226 dose. All deaths were attributed to disease progression.
original articles
Annals of Oncology
Table 2. Summary of dose-limiting toxic effects (DLTs) occurring during the first BGT226 treatment cycle BGT226 dose cohort
DLT
Description
62-year-old female with ovarian carcinoma
125 mg/day, TIW
Diarrhea and vomiting (grade 3)
43-year-old male with schwannoma
125 mg/day, TIW
Diarrhea (grade 3)
59-year-old female with breast cancer
125 mg/day, TIW
Vomiting (grade 3)
C1D1—Patient had CTCAE grade 1 vomiting post dose despite predose of palonosetron. C1D3—Patient received palonosetron and dexamethasone as premedication and still suffered post dose CTCAE grade 1 vomiting and CTCAE grade 2 diarrhea. C1D6—Patient was premedicated with prochlorperazine and dexamethasone and experienced CTCAE grade 3 vomiting and diarrhea ∼5 h after dose. Vomiting was treated with prochlorperazine, and diarrhea was treated with lomotil. Both diarrhea and vomiting were considered as a DLT. C1D8—Patient discontinued study medication due to the adverse events suffered. C1—In the fourth week patient had diarrhea CTCAE grade 3, ∼4 h after dosing of BGT226. Despite optimal treatment of loperamide, patient had seven episodes of diarrhea (CTCAE grade 3) on two treatment days. This event was assessed as DLT, and the dose was reduced to 80 mg. Following dose reduction, the patient continued to experience CTCAE grade 1 diarrhea despite optimal treatment with loperamide. C1D8—Patient experienced CTCAE grade 1 vomiting post dose that resolved with palonosetron hydrochloride and lorazepam treatment. C1D15—Patient was premedicated with prochlorperazine. Post dose patient developed nausea/vomiting (CTCAE grade 3) and diarrhea (CTCAE grade 2). Patient was treated with lorazepam, prochlorperazine, lomotil, and prednisone. Diarrhea was not maximally treated. Vomiting was assessed as DLT, as it did not resolve with standard antiemetic treatment.
C, cycle; CTCAE, Common Terminology Criteria for Adverse Events; D, day; TIW, three times weekly.
Table 3. Patient exposure to BGT226, dose modifications, and disposition BGT226 dose cohort (mg/day) 2.5 (N = 5) 5 (N = 4) 10 (N = 4) 20 (N = 3) 40 (N = 8) 80 (N = 9) 100 (N = 13) 125 (N = 11) All (N = 57) Exposure, n (%) ≤ 4 weeks 2 (40) > 4 to ≤8 weeks 2 (40) > 8 to ≤12 weeks 0 > 12 weeks 1 (20) Median exposure, weeks 4.1 Minimum 1.1 Maximum 37.7 Patients with one or more dose reduction, n (%) 0 Due to AE 0 Patients with one or more dose delay due to AE, n (%) 2 (40) Treatment discontinued, n (%) 5 (100) Primary reason for treatment discontinuation, n (%) AE 0 Disease progression 5 (100) Consent withdrawal 0
1 (25) 1 (25) 1 (25) 1 (25) 8.0 3.7 31.7 0 0 0 4 (100)
1 (25) 2 (50) 0 1 (25) 5.5 2.4 19.7 0 0 0 4 (100)
0 0 4 (100) 2 (50) 0 2 (50)
1 (33) 2 (67) 0 0 4.1 3.9 7.9 0 0 0 3 (100)
1 (13) 4 (50) 2 (25) 1 (13) 7.7 0.9 15.9 0 0 1 (13) 8 (100)
1 (11) 4 (44) 0 4 (44) 7.7 3.4 31.6 1 (11) 0 3 (33) 9 (100)
3 (23) 4 (31) 3 (23) 3 (23) 7.9 3.7 48.1 1 (8) 0 6 (46) 13 (100)
3 (27) 5 (45) 2 (18) 1 (9) 7.6 0.9 12.3 4 (36) 4 (36) 7 (64) 11 (100)
13 (23) 24 (42) 8 (14) 12 (21) 7.6 0.9 48.1 6 (11) 4 (7) 19 (33) 57 (100)
0 3 (100) 0
0 8 (100) 0
1 (11) 8 (89) 0
0 13 (100) 0
3 (27) 8 (73) 0
4 (7) 51 (90) 2 (4)
Data cut-off 10 February 2010. AE, adverse event.
high and ranged (CV%) from 50% to 100% and from 30% to 200%, respectively, for most cohorts. Systemic exposure to BGT226 increased with increasing dose (Figure 1). It was, however, generally lower than expected based on preclinical data, and this finding was also supported by lower than
| Markman et al.
expected Cmax values, particularly at the 100 mg/day dose (Table 5). The median terminal elimination half-life ranged from 6 to 9 h after the first dose. No significant drug accumulation in plasma was reported following repeat administration.
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Patient
original articles
Annals of Oncology Table 4. Incidence of adverse events by BGT226 treatment cohort
BGT226 dose cohorts (mg/day) 2.5 (N = 5) 5 (N = 4) 10 (N = 4) 20 (N = 3) 40 (N = 8) 80 (N = 9) 100 (N = 13) 125 (N = 11) All (N = 57) All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 12 (92) 12 (92) 10 (77) 8 (62) 6 (46) 2 (15) 4 (31)
0 0 0 1 (8) 1 (8) 1 (8) 0
10 (91) 10 (91) 9 (82) 3 (27) 3 (27) 3 (27) 2 (18)
2 (18) 5 (46) 2 (18) 1 (9) 0 0 0
45 (79) 39 (68) 33 (58) 19 (33) 19 (33) 14 (25) 12 (21)
2 (4) 7 (12) 2 (4) 3 (5) 1 (2) 4 (7) 1 (2)
12 (92) 10 (77) 10 (77) 4 (31) 5 (39) 2 (15) 0
0 0 0 0 1 (8) 1 (8) 0
9 (82) 10 (91) 8 (73) 1 (9) 0 1 (9) 1 (9)
2 (18) 5 (46) 2 (18) 0 0 0 0
39 (68) 35 (61) 28 (49) 11 (19) 9 (16) 6 (11) 5 (9)
2 (4) 7 (12) 2 (4) 0 1 (2) 1 (2) 0
Gr, grade.
antitumor activity Stable disease (SD), defined as at least one SD assessment after start of treatment, was the best overall response achieved with BGT226 treatment. Response outcome as per investigator review is available for 51 patients; for 6 patients, there was either no documentation of progression or one or more target lesions were not assessed or were assessed using a different method than baseline. Seventeen of 57 patients (30%) achieved SD as best response. Tumor shrinkage, not qualifying for partial response, was observed in three of 57 patients (two with SD, one unknown best response). Nine patients (16%) had a progression-free period of four or more cycles. Thirty-four patients (60%) had progressive disease. Thirty patients (53%) achieved stable metabolic disease (SMD) as best metabolic response by central review of FDG– PET imaging. Metabolic progressive disease was reported in 10 patients (18%), and in the remaining patients, the percentage change from baseline in sSUVmax could not be assessed. Baseline and cycle 1 day 28 FDG–PET assessments were available for 38 patients (Figure 2). There was no apparent correlation between metabolic response by FDG–PET and standard radiological outcome measures determined by computed tomography (CT).
PD analysis Paired skin biopsies were available for 49 patients; 22 patients demonstrated > 25% decrease in pS6 levels at cycle 1, day 8 and day 28 compared with baseline (Figure 3A). However, consistent reduction in pS6 was present in relatively few cases, mostly in the 125-mg cohort, in which four of eight patients achieved reductions in pS6 (range 9% − 64% decrease) at both posttreatment time points; reduction was more pronounced
Volume 23 | No. 9 | September 2012
(range 37% − 64% decrease) after prolonged exposure to BGT226. A limited number of paired tumor biopsies were available (2.5-mg cohort, N = 3; 10-, 20-, and 40-mg cohorts, N = 1 each). Reductions in pS6 and pAkt levels, indicating inhibition of the PI3K pathway, were observed in tumor biopsies from the 2.5-mg and 20-mg cohorts, one of which (at 20 mg) also demonstrated a change in cellular proliferation (Figure 3B, C and D). Approximately half of the patients showed an increase in circulating cleaved and intact CK18, as measured by M30 (indicating epithelial cell apoptosis) and M65 (any type of cell death). Increases in total CK18 levels (M65) occurred mostly at BGT226 doses of 80 to 125 mg (data not shown). Changes in the circulating angiogenesis markers VEGF, sVEGFR1/2, PIGF, and bFGF, although evident in some patients, did not correlate with BGT226 dose levels (data not shown).
discussion In this first-in-human study, the MTD of the oral PI3K/mTOR inhibitor BGT226 in patients with advanced solid tumors was 125 mg/day, TIW; the recommended dose was defined as 100 mg administered once a day, TIW. The majority of patients (88%) experienced AEs potentially related to BGT226, with gastrointestinal effects and fatigue being the most frequent, consistent with preclinical observations. No grade >1 abnormalities in FPG were observed, in contrast to the inhibition of glucose metabolism and hyperglycemia associated with PI3K inhibition in BGT226 preclinical studies. The best overall response achieved with BGT226 was SD in 17 (30%) of patients. Tumor shrinkage not qualifying for partial response was observed in 3 of 57 patients (two with SD,
doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Most common adverse events, occurring in > 15% of patients, regardless of study drug relationship, n (%) Nausea 3 (60) 0 2 (50) 0 1 (25) 0 3 (100) 0 6 (75) 0 8 (89) 0 Diarrhea 1 (20) 0 1 (25) 0 1 (25) 0 2 (67) 0 4 (50) 0 8 (89) 2 (22) Vomiting 2 (40) 0 0 0 2 (50) 0 1 (33) 0 4 (50) 0 5 (56) 0 Decreased appetite 1 (20) 1 (20) 1 (25) 0 2 (50) 0 2 (67) 0 1 (13) 0 1 (11) 0 Fatigue 1 (20) 0 3 (75) 0 0 0 0 0 1 (13) 0 5 (56) 0 Anemia 2 (40) 2 (40) 1 (25) 0 1 (25) 1 (25) 1 (33) 0 2 (25) 0 2 (22) 0 Abdominal pain 1 (20) 1 (20) 1 (25) 0 1 (25) 0 1 (33) 0 1 (13) 0 1 (11) 0 Adverse events suspected to be related to study drug occurring in > 2 (4%) of patients, n (%) Nausea 1 (20) 0 1 (25) 0 1 (25) 0 2 (67) 0 5 (63) 0 8 (89) 0 Diarrhea 0 0 1 (25) 0 1 (25) 0 2 (67) 0 3 (38) 0 8 (89) 2 (22) Vomiting 0 0 0 0 1 (25) 0 1 (33) 0 3 (38) 0 5 (56) 0 Fatigue 0 0 0 0 0 0 0 0 1 (13) 0 5 (56) 0 Decreased appetite 0 0 0 0 2 (50) 0 1 (33) 0 0 0 1 (11) 0 Anemia 0 0 0 0 0 0 0 0 1 (13) 0 2 (22) 0 Asthenia 1 (20) 0 1 (25) 0 0 0 1 (33) 0 0 0 1 (11) 0
original articles
Annals of Oncology
Table 5. Summary of BGT226 pharmacokinetic parameters by cohort PK parameter (unit)
2.5 mg (N = 5)
AUC(0–48) (h*ng/ml) Cmax (ng/ml) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h)
5 mg (N = 4)
10 mg (N = 4)
20 mg (N = 3)
40 mg (N = 8)
80 mg (N = 9)
100 mg (N = 13)
125 mg (N = 11)
Mean Mean (SD) Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median
Cycle 1, day 1
Cycle 1, day 8
Cycle 2, day 1
— 0.18 (0.117) 1.53 1.73 (0.166) 0.42 (0.162) 9.22 1.09 4.45 (3.501) 0.81 (0.863) 6.31 2.55 9.79 (3.745) 1.47 (0.460) 8.35 2.00 11.67 (3.944) 1.47 (0.764) 7.76 3.00 32.86 (20.071) 2.50 (0.938) 7.61 3.00 26.97 (17.752) 2.45 (1.869) 8.61 3.00 78.75 (50.548) 5.21 (2.571) 8.62 4.00
0.49 0.11 (0.063) 0.97 2.29 (0.657) 0.31 (0.146) 17.74 1.99 5.37 (4.623) 0.78 (0.641) 8.79 2.25 14.41 (8.639) 1.86 (1.018) 7.83 3.12 12.12 (4.744) 0.92 (0.533) 8.86 2.48 32.58 (20.827) 2.70 (1.464) 8.03 3.00 32.11 (29.674) 2.59 (2.330) 9.84 3.00 86.66 (80.831) 4.40 (2.540) 7.09 3.95
— 0.13 (0.104) 2.50 1.49 (0.274) 0.34 (0.225) 13.80 0.97 3.62 (2.504) 0.38 (0.232) 9.98 2.99 15.35 (18.362) 1.67 (1.677) 7.64 4.61 12.07 (4.012) 1.47 (0.449) 11.45 2.26 30.52 (22.125) 3.07 (2.895) 9.16 3.00 26.41 (17.968) 2.43 (2.080) 8.18 2.00 68.89 (55.592) 4.14 (3.438) 11.10 4.54
Samples were collected predose and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 24, 32, and 48 h post dose. Not all patients in each cohort were assessable for all parameters for all time points. AUC, area under the plasma concentration versus time curve from 0 to 48 h (AUC0–48); Cmax, maximum plasma concentration; PK, pharmacokinetics; SD, standard deviation; Tmax, time to Cmax; t1/2, elimination half-life. a
one unknown). Nine patients (16%) remained progression free for ≥16 weeks. Thirty patients (53%) had SMD on FDG–PET evaluations; however, there was no apparent correlation between metabolic response and tumor shrinkage according to CT. It should be noted that the role of PI3K signaling in glucose uptake and metabolism [1] may confound interpretation of FDG–PET findings following treatment with BGT226, as detectable changes may reflect PI3K pathway downregulation more than antitumor activity. Changes in FDG–PET imaging with this class of agent may therefore have a PD component before indicating an antitumor effect [2]; however, further study is required. The PK profile of BGT226 was characterized by rapid absorption. The onset of vomiting on dosing days appeared to be correlated with Tmax/Cmax. Systemic exposure to BGT226 generally increased with increasing dose, and there was no significant accumulation of BGT226 following repeat dosing. In general, plasma levels of BGT226 were found to be 25- to
| Markman et al.
100-fold lower than expected based on preclinical data, which may have resulted in PK variability, suboptimal drug levels, and limited PD and antitumor activity. PD analyses suggested evidence of PI3K pathway inhibition following BGT226 treatment in selected patients. Overall, however, the PD effects of BGT226 were inconsistent. At BGT226 doses >80 mg, PI3K pathway inhibition was more evident in skin biopsies from a few patients, as determined by a reduction in pS6 (Ser240/244) by 37% − 64%. Available data suggest that BGT226 may also inhibit the PI3K pathway ( pAkt and pS6) in tumors; however, reductions of pS6/pAkt in tumors were not seen in all samples, and there were no available biopsies at doses ≥80 mg, making further interpretation difficult. These findings suggest that there was insufficient inhibition of the PI3K pathway in tumors by BGT226 to translate into significant antitumor effects with the explored dosing schedule. Consistently, no objective responses were observed with BGT226. Modeling based on PK data from this study and subsequent preclinical PK/PD analysis predicted
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
BGT226 dose cohorta
Annals of Oncology
original articles
that BGT226 doses >4000 mg/day would be required to achieve efficacious plasma exposure, greatly exceeding the human safety dose range. The four class I PI3K isoforms have overlapping nonredundant roles in cellular signal transduction [16]; however, preclinical data suggest context-specific functions. p110α is frequently mutated or amplified in human cancers, whereas overexpression of wild-type p110 β, δ, and γ isoforms can induce neoplastic transformation in cultured cells [3]. The PI3K inhibitors in clinical development achieve differing potencies against these isoforms, leading to speculation that the inhibitory profile of any given compound may have a significant impact on tolerability and activity. Preliminary clinical evidence suggests that this may be the case. Recent phase I study results of the novel dual PI3K/mTOR inhibitor BEZ235 in patients with advanced solid tumors showed that BEZ235 was well tolerated with a favorable safety profile and manageable gastrointestinal side-effects [17]. Notably, in the BEZ235 study, there were two partial responses in a patient with breast cancer and in a patient with lung cancer and Cowden syndrome [2]. In addition, recent results from a phase I study of the selective pan-class I PI3K inhibitor BKM120 also demonstrated a favorable safety and activity profile, including a partial response in a patient with breast cancer [18]. Preclinical data strongly support the use of PI3K inhibitors in tumors with PI3K pathway genetic alterations [9, 10, 19, 20]. Originally, it was intended in this study to investigate the relation between tumor mutational status and tumor response. However, as no objective responses were observed with BGT226, this comparison is obsolete and no conclusion to guide patient selection based on molecular tumor status can be drawn from this study. This may be due to insufficient
Volume 23 | No. 9 | September 2012
exposure, although the role of cellular signaling cross talk/ feedback (i.e. via the Ras pathway) cannot be excluded. It should be noted that patients both with and without identified tumor PI3K pathway alterations have demonstrated responses to other PI3K pathway inhibitors, including BEZ235 and BKM120 [17, 18]. BGT226 is one of the first compounds to be evaluated in the emerging PI3K inhibitor clinical development program. Although a small number of patients in our study derived some clinical benefit from BGT226, overall, the data suggest that it is not feasible to consistently achieve clinically significant antitumor effects with BGT226 at safely administered doses. Despite these findings, these data are valuable for ongoing and future studies of other PI3K-targeted inhibitors.
acknowledgements We thank the following investigators, Steven Isakoff (Massachussets General Hospital, Boston, USA), Sunil Sharma (Nevada Cancer Institute, Las Vegas, USA), Christophe LeTourneau (Princess Margaret Hospital, Toronto, Canada), and contributors Beatrix Staffler, Humphrey Gardner, Nina Huseinovic, Nassim Sleiman, Mikael Sacco, Hartmut Zehender, and Stefan de Buck (Novartis Pharma AG), whose efforts were essential toward the completion of this work. We would also like to thank the European Society of Medical Oncology (ESMO) for awarding BM with an ESMO fellowship grant that contributed to his involvement in this trial. The authors would also like to acknowledge Dina Marenstein, PhD, of Chameleon Communications International, who provided medical writing services with funding from Novartis Pharmaceuticals Corporation.
doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 1. Plasma pharmacokinetic profile of BGT226 in patients with solid tumors. Mean plasma area under the curve (AUCinf ) values are shown for each dose cohort for cycle 1, day 1 and day 8, and cycle 2, day 1 time points.
original articles
Annals of Oncology
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 2. (A) Best percentage change from baseline in sum of longest diameters for 42 patients, best overall response as per central review by indication. (B) Tumor metabolic response to BGT226 therapy by [18F]-fluorodeoxyglucose–positron emission tomography (FDG–PET). Best percentage change from baseline to cycle 1, day 28 in the sum of maximum standardized uptake values (sSUVmax) for 38 patients with baseline and cycle 1, day 28 assessments.
| Markman et al.
Volume 23 | No. 9 | September 2012
Annals of Oncology
original articles
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 3. Pharmacodynamic effects of BGT226 treatment in skin and tumor biopsies. (A) Effect of BGT226 treatment on pS6 levels in skin biopsies. Data are presented as percent change from baseline of the H-score of the whole epidermis for evaluable posttreatment (day 8 and day 28 of cycle 1) samples by dose cohort. Effect of BGT226 treatment on (B) pS6, (C) pAkt, and (D) Ki-67 levels in tumor biopsies from a patient with colon cancer treated with 20 mg BGT226 ( percent change from baseline to cycle 2, day 1 is shown).
Volume 23 | No. 9 | September 2012
doi:10.1093/annonc/mds011 |
original articles funding Novartis Pharmaceuticals Corporation.
disclosure MMC, SS, DS, OA, and WH are employees of Novartis Pharma AG, the developer of BGT226 and the study sponsor. IK receives clinical trial funding from Novartis Pharma AG. JB has received consulting honoraria from Novartis. BM, JT, GIS, LS, LCC, and MM have no conflict of interest to declare.
| Markman et al.
10. Serra V, Markman B, Scaltriti M et al. NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Res 2008; 68: 8022–8030. 11. Maira SM, Voliva C, Garcia-Echeverria C. Class IA phosphatidylinositol 3-kinase: from their biologic implication in human cancers to drug discovery. Expert Opin Ther Targets 2008; 12: 223–238. 12. Maira SM, Stauffer F, Brueggen J et al. Identification and characterization of NVP-BEZ235, a new orally available dual phosphatidylinositol 3-kinase/ mammalian target of rapamycin inhibitor with potent in vivo antitumor activity. Mol Cancer Ther 2008; 7: 1851–1863. 13. Neuenschwander B, Branson M, Gsponer T. Critical aspects of the Bayesian approach to phase I cancer trials. Stat Med 2008; 27: 2420–2439. 14. Babb J, Rogatko A, Zacks S. Cancer phase I clinical trials: efficient dose escalation with overdose control. Stat Med 1998; 17: 1103–1120. 15. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 16. Engelman JA, Luo J, Cantley LC. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 2006; 7: 606–619. 17. Burris H, Rodon J, Sharma S et al. First-in-human phase I study of the oral PI3K inhibitor BEZ235 in patients with advanced solid tumors. Presented at the 2010 American Society of Clinical Oncology Annual Meeting (Abstr 3005). Chicago, IL June 4–8, 2010. 18. Bendell JC, Rodon J, Burris HA et al. Phase I, dose-escalation study of BKM120, an oral pan-class I PI3K inhibitor, in patients with advanced solid tumors. J Clin Oncol 2012; 30: 282–290. 19. Ihle NT, Lemos R, Jr, Wipf P et al. Mutations in the phosphatidylinositol-3-kinase pathway predict for antitumor activity of the inhibitor PX-866 whereas oncogenic Ras is a dominant predictor for resistance. Cancer Res 2009; 69: 143–150. 20. Engelman JA, Chen L, Tan X et al. Effective use of PI3K and MEK inhibitors to treat mutant Kras G12D and PIK3CA H1047R murine lung cancers. Nat Med 2008; 14: 1351–1356.
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
1. Katso R, Okkenhaug K, Ahmadi K et al. Cellular function of phosphoinositide 3kinases: implications for development, homeostasis, and cancer. Annu Rev Cell Dev Biol 2001; 17: 615–675. 2. Engelman JA. Targeting PI3K signalling in cancer: opportunities, challenges and limitations. Nat Rev Cancer 2009; 9: 550–562. 3. Courtney KD, Corcoran RB, Engelman JA. The PI3K pathway as drug target in human cancer. J Clin Oncol 2010; 28: 1075–1083. 4. Liu P, Cheng H, Roberts TM et al. Targeting the phosphoinositide 3-kinase pathway in cancer. Nat Rev Drug Discov 2009; 8: 627–644. 5. Vogt PK, Kang S, Elsliger MA et al. Cancer-specific mutations in phosphatidylinositol 3-kinase. Trends Biochem Sci 2007; 32: 342–349. 6. Brognard J, Clark AS, Ni Y et al. Akt/protein kinase B is constitutively active in non-small cell lung cancer cells and promotes cellular survival and resistance to chemotherapy and radiation. Cancer Res 2001; 61: 3986–3997. 7. Berns K, Horlings HM, Hennessy BT et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell 2007; 12: 395–402. 8. Hu L, Hofmann J, Lu Y et al. Inhibition of phosphatidylinositol 3’-kinase increases efficacy of paclitaxel in in vitro and in vivo ovarian cancer models. Cancer Res 2002; 62: 1087–1092. 9. Eichhorn PJ, Gili M, Scaltriti M et al. Phosphatidylinositol 3-kinase hyperactivation results in lapatinib resistance that is reversed by the mTOR/phosphatidylinositol 3kinase inhibitor NVP-BEZ235. Cancer Res 2008; 68: 9221–9230.
Annals of Oncology
Annals of Oncology
Annals of Oncology 23: 2399–2408, 2012 doi:10.1093/annonc/mds011 Published online 22 February 2012
Phase I safety, pharmacokinetic, and pharmacodynamic study of the oral phosphatidylinositol-3-kinase and mTOR inhibitor BGT226 in patients with advanced solid tumors B. Markman1,†, J. Tabernero1, I. Krop2, G. I. Shapiro2, L. Siu3, L. C. Chen4, M. Mita5, M. Melendez Cuero6, S. Stutvoet6, D. Birle7, Ö. Anak6,‡, W. Hackl6 & J. Baselga1, §*
Received 7 June 2011; revised 22 November 2011; accepted 10 January 2012
Background: This phase I dose-escalation study investigated the maximum tolerated dose (MTD), safety, pharmacokinetics, pharmacodynamics (PDs), and preliminary antitumor activity of BGT226, a potent, oral dual phosphatidylinositol-3-kinase (PI3K)/mammalian target of rapamycin inhibitor. Patients and methods: Fifty-seven patients with advanced solid tumors received BGT226 2.5–125 mg/day three times weekly (TIW). Dose escalation was guided by an adaptive Bayesian logistic regression model with overdose control. Assessments included response per RECIST, [18F]-fluorodeoxyglucose uptake, and phosphorylated-S6 in skin and paired tumor samples. Results: Three patients (125 mg cohort) had dose-limiting toxic effects (grade 3 nausea/vomiting, diarrhea). BGT226related adverse events included nausea (68%), diarrhea (61%), vomiting (49%), and fatigue (19%). BGT226 demonstrated rapid absorption, variable systemic exposure, and a median half-life of 6–9 h. Seventeen patients (30%) had stable disease (SD) as best response. Nine patients had SD for ≥16 weeks. Thirty patients (53%) achieved stable metabolic disease as assessed by [18F]-fluorodeoxyglucose–positron emission tomography; however, no correlation between metabolic response and tumor shrinkage according to computed tomography was observed. PD changes suggested PI3K pathway inhibition but were inconsistent. Conclusions: The MTD of BGT226 was 125 mg/day TIW, and the clinically recommended dose was 100 mg/day TIW. Limited preliminary antitumor activity and inconsistent target inhibition were observed, potentially due to low systemic exposure. Key words: anticancer agent, BGT226, dual inhibitor, mTOR catalytic inhibitor, PI3K pathway, solid tumors
introduction The phosphatidylinositol-3-kinase (PI3K) signal transduction pathway regulates key physiologic functions, including cellular proliferation, growth, metabolism, and survival [1, 2]. It is *Correspondence to: Dr J. Baselga, Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Massachusetts General Hospital, 55 Fruit Street, Lawrence House 108, Boston, MA 02114, USA. Tel: +1-617-643-2438; Fax: +1-617-643-2683; E-mail: [email protected] †
Present address: Monash Institute of Medical Research, Monash University, Melbourne, Australia.
‡
Present address: Oncology Global Development, Novartis Pharma AG, Basel, Switzerland.
§
Present address: Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center, Boston, USA.
frequently activated in cancer cells, via several mechanisms, including mutation of PI3K signaling components and downstream effectors [3, 4]. Activating mutations in PIK3CA, which encodes the p110α catalytic subunit of PI3K, are present in many human cancers, including breast, colon, endometrial, and hepatocellular cancers [3–5]. Evidence suggests that PI3K pathway activation is oncogenic and may contribute to therapeutic resistance to chemotherapy and targeted agents in a number of cancers, including breast and lung cancer [3, 6–9]. These data suggest that inhibition of PI3K signaling is an attractive target for anticancer therapy, and preclinical studies of PI3K signaling inhibitors demonstrated antiproliferative and antitumor effects [3, 10–12]. BGT226 (Novartis Pharma AG, Basel, Switzerland), an imidazoquinoline derivative, is an orally bioavailable potent
© The Author 2012. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
1 Medical Oncology Department, Vall d’Hebron University Hospital, Universitat Autònoma de Barcelona, Barcelona, Spain; 2Department of Medical Oncology, DanaFarber Cancer Institute, Boston, USA; 3Department of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Canada; 4Nevada Cancer Institute, Las Vegas; 5Cancer Therapy and Research Center, The University of Texas Health Science Center at San Antonio, San Antonio, USA; 6Oncology Translational Medicine, Novartis Pharma AG, Basel, Switzerland; 7Novartis Institutes for Biomedical Research, Cambridge, USA
original articles
patients and methods See supplemental Appendix (available at Annals of Oncology online) for additional details.
The starting dose of 2.5 mg PO on a ‘q48h’ schedule is based on in vivo experiments in rat and dog, which showed a long-lasting effect of BGT226 and a similar efficacy to a q24h schedule (see supplemental Appendix, available at Annals of Oncology online, for additional details). Dose escalation was guided by an adaptive Bayesian logistic regression model with overdose control and proceeded by enrolling cohorts of three or more patients [13, 14]. The dose-limiting toxicity (DLT) rate was evaluated after the first treatment cycle. The dose of the subsequent cohort was chosen to fulfill the overdose criterion of a < 25% chance of the DLT rate exceeding 33%. Dose escalation was further based on review of clinical safety, PK, and laboratory data. DLT was defined as a suspected drug-related toxicity according to National Cancer Institute—Common Terminology Criteria for Adverse Events (NCI–CTCAE) V3.0, unless otherwise specified, occurring < 28 days following the first BGT226 dose. The MTD was defined as the highest dose resulting in DLTs in ≤ 33% of patients in cycle 1.
safety and response assessments Safety assessments were conducted at baseline and weekly during the first eight treatment weeks and every 2 weeks thereafter. Adverse events (AEs) were graded using NCI–CTCAE, unless otherwise specified (supplemental Table S1, available at Annals of Oncology online). All patients had medical history, physical examination, vital signs, hematology and chemistry profile, cardiac enzymes, and electrocardiogram (ECG) assessments. Fasting serum insulin, serum C-peptide, and fasting plasma glucose (FPG) were analyzed weekly during cycle 1 and on days 15 and 28 in subsequent cycles. Hemoglobin A1c and fructosamine were measured at baseline and day 1 of subsequent cycles, starting with cycle 2. Urine glucose was monitored daily in cycle 1. Response was assessed according to the RECIST V1.0 [15] every 8 weeks by investigator and central review. The metabolic antitumor activity of BGT226 was assessed by [18F]fluorodeoxyglucose (FDG)–positron emission tomography (PET), evaluating change from baseline in the sum of maximum standardized uptake values (sSUVmax) in sequential PET scans carried out at baseline and day 28 of cycle 1.
study design The primary objective of this multicentre, open-label phase I doseescalation study was to determine the MTD of oral single-agent BGT226 administered to adult patients with solid tumors. Secondary objectives included evaluation of safety and tolerability, measurement of PK end points, assessment of antitumor activity, and PD effects. Main inclusion criteria were histologically confirmed advanced solid tumors unresponsive to standard therapy, age ≥18 years, life expectancy ≥12 weeks, World Health Organization performance status two or less, and adequate bone marrow, hepatic, and renal function. Key exclusion criteria included prior treatment with PI3K inhibitors, diabetes mellitus, brain metastases, gastrointestinal disease that could affect BGT226 absorption, progressive eye disease, thyroid disease requiring treatment, acute or chronic liver or renal disease, or pancreatitis. The study was conducted in accordance with the International Conference on Harmonisation Good Clinical Practice Guidelines and the Declaration of Helsinki, with approval by ethics committees and health authorities of participating institutions. All patients provided written informed consent.
treatment and dose-escalation criteria Patients were instructed to take BGT226 (hard gelatin capsules) as a single oral daily dose, three times weekly (TIW) (every other day), 2 h after a light breakfast, and to continue fasting for two more hours. Treatment was continued on 28-day treatment cycles until unacceptable toxicity, disease progression, or consent withdrawal.
| Markman et al.
PK analysis Plasma BGT226 levels were measured on days 1 and 8 of cycle 1, day 1 of cycle 2, predose, and at indicated time points post dose using a validated liquid chromatography–tandem mass spectrometry assay with a lower quantification limit of 10 pg/ml. PK parameters were calculated by noncompartmental analysis using WinNonlin® (Pharsight, Mountain View, CA).
PD assessments Immunohistochemical analysis of phosphorylated Akt (serine 473; pAkt), phosphorylated ribosomal protein S6 (serine 240/244; pS6), and Ki-67 were carried out in available pre- and posttreatment skin biopsies and optional paired fresh pre- and posttreatment tumor biopsies. Blood samples were analyzed for levels of circulating angiogenic markers and markers of cell death.
results Fifty-seven patients across five sites were treated with BGT226 between December 2007 and February 2010 (Table 1). Breast (23%), prostate (14%), and colon cancers (12%) were the most common malignancies. All patients were evaluated for safety, response, and PK.
dose escalation and safety findings Patients were enrolled into sequential cohorts at BGT226 dose levels of 2.5, 5, 10, 20, 40, 80, and 125 mg. In the 125-mg
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
pan-class I PI3K and mammalian target of rapamycin (mTOR) catalytic inhibitor. BGT226 is among the first in a novel class of agents targeting the PI3K pathway. BGT226 specifically inhibits p110α, β, δ, and γ, with a preference for the α-isoform (wild type and mutated), and mTOR (Novartis, data on file), with no significant inhibitory activity against other tested kinases (see supplemental Appendix, available at Annals of Oncology online). The IC50 of BGT226 for the individual isoforms of the class I PI3Ks is (±standard deviation) PI3Kα 4 nM (±1), PI3Kβ 63 nM (±10), and PI3Kγ 38 nM (±23) as determined by filter-binding assay (Novartis, data on file). In preclinical studies, BGT226 inhibited cancer cell proliferation with IC50 values in the low nanomolar range and inhibited solid tumor growth in various mouse xenograft models, including glioblastoma multiforme, breast, and prostate cancer (Novartis, data on file). In vivo pharmacokinetic (PK)/ pharmacodynamic (PD) analyses showed a good correlation between BGT226 dose, blood, and tissue levels, effect on the PI3K pathway (determined by Akt phosphorylation), and antitumor activity (Novartis, data on file). Here, we report the results of a phase I, first-in-human doseescalation study of BGT226 to determine the maximum tolerated dose (MTD) when administered orally (PO) to adults with advanced solid tumors. Safety, tolerability, PK/PD properties, and preliminary clinical activity of BGT226 were also evaluated.
Annals of Oncology
original articles
Annals of Oncology Table 1. Baseline patient characteristics Characteristics
60.0 23–82 24 (42.1) 33 (57.9) 28 (49.1) 26 (45.6) 3 (5.3) 13 (22.8) 8 (14.0) 7 (12.3) 4 (7.0) 4 (7.0) 3 (5.3) 2 (3.5) 2 (3.5) 2 (3.5) 12 (21.1) 1.4 1–8 1 (1.8) 56 (98.2) 57 (100) 4 1–13 32 (56.1)
a
One patient each had bladder, cervix, cholangiocarcinoma, kidney, fallopian tubes, mucosal melanoma, pancreas, mesothelioma, rectum, salivary gland, and stomach cancer, and gastrointestinal stromal tumor. ECOG, Eastern Cooperative Oncology Group; WHO, World Health Organization.
cohort, 3 of 11 patients had DLTs, all gastrointestinal in nature: grade 3 diarrhea, grade 3 vomiting, and grade 3 diarrhea and vomiting (Table 2). Based on the DLT incidence, the MTD of BGT226 was defined at 125 mg/day, TIW. As a result of the gastrointestinal AEs associated with the 125 mg dose, a deescalated 100 mg dose level was explored, which resulted in improved tolerance. BGT226 100 mg/day, TIW, was therefore declared the clinically recommended dose. Overall, patients were exposed to BGT226 for a median of 7.6 weeks (Table 3). As of 10 February 2010, all patients had discontinued treatment. Four patients (7%) discontinued due to AEs; two of these were considered treatment related (diarrhea with vomiting, one patient and diarrhea with nausea/ vomiting, one patient, both in the 125-mg cohort), and two were assessed as nontreatment related (sepsis of urological origin, one patient, 125-mg cohort and pathological fracture, one patient, 80-mg cohort). Twelve patients (21%) required dose adjustment or interruption due to AEs.
Volume 23 | No. 9 | September 2012
pharmacokinetics BGT226 was rapidly absorbed after single-dose oral administration, with median peak plasma concentrations (Cmax) observed between 1.0 and 4.6 h post dose (Tmax) (Table 5). Interpatient variability in Cmax and area under the plasma concentration time curve (AUC0 − 48) was relatively doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Age, years Median Range Sex, n (%) Male Female WHO/ECOG performance status, n (%) 0 1 2 Primary tumor, n (%) Breast Prostate Colon Soft tissue sarcoma Uterus Bone Ovary Esophagus Skin melanoma Othera Time since most recent recurrence/relapse, months Median Range Current stage, n (%) Stage III Stage IV Prior antineoplastic therapy, n (%) Number of regimens Median Range Patients with more than three prior regimens, n (%)
No. of patients (N = 57)
Gastrointestinal disorders were the most frequently reported AEs regardless of study drug relationship and included nausea, diarrhea, vomiting, decreased appetite, and abdominal pain (Table 4). Other common AEs included fatigue (33%) and anemia (25%). Other hematological abnormalities were less common (lymphopenia, 10.5%; thrombocytopenia, 3.5%; and neutropenia, 1.8%). Fifty patients (88%) experienced suspected treatment-related AEs. The most common were nausea (68%), diarrhea (61%), vomiting (49%), fatigue (19%), and anorexia (16%) (Table 4). Eleven patients (19%) experienced grade 3/4 drug-related AEs, of which diarrhea was the most common (12%). The other frequent, suspected drug-related AEs were generally mild to moderate in severity. The incidence of drug-related vomiting, nausea, and diarrhea was dose dependent and generally limited to dosing days. Only after the first onset of vomiting, nausea, and/or diarrhea, patients were pretreated and/or treated with antiemetics and/or antidiarrheals, respectively, according to the site’s standards. Medications used included, among others, loperamide and lomotil for diarrhea and aprepitant, metoclopramide, palonosetron, lorazepam, and prochlorperazine for vomiting and nausea. Despite pretreatment and/or treatment of these events, the 125 mg dose was not well tolerated. Generally, the 100 mg dose was well tolerated due to these measures. The only hematological toxicity potentially related to BGT226 was anemia (11%). Grade 3 anemia regardless of relationship was observed in four patients (two in the 2.5-mg cohort, one each in the 10-mg and 100-mg cohorts). For these patients, grade 1 or 2 anemia was already noted at screening and/or baseline, and hemoglobin was low throughout the study. Most changes in biochemical parameters were mild in severity (grade 1). The most frequent grade 2 or higher biochemical abnormalities were aspartate aminotransferase (grade 2, N = 9; grade 3, N = 3) and alkaline phosphatase (grade 2, N = 4; grade 3, N = 4) elevations. Two patients had newly occurring or grade 4 hyperuricemia. No other grade 4 abnormalities of biochemistry parameters were noted. Two patients had grade 1 new or worsened FPG (10-mg and 100mg cohorts). No grade > 1 FPG abnormality was observed in any patient. No clinically significant abnormalities were observed by cardiac imaging or ECG after treatment initiation. Serious AEs (SAEs) were experienced by 10 patients (17.5%) (supplemental Table S2, available at Annals of Oncology online). One patient (125-mg cohort) experienced SAEs considered by investigators to be treatment related (diarrhea, nausea, and vomiting). There were no deaths on BGT226 treatment. Four patients died within 28 days after the last BGT226 dose. All deaths were attributed to disease progression.
original articles
Annals of Oncology
Table 2. Summary of dose-limiting toxic effects (DLTs) occurring during the first BGT226 treatment cycle BGT226 dose cohort
DLT
Description
62-year-old female with ovarian carcinoma
125 mg/day, TIW
Diarrhea and vomiting (grade 3)
43-year-old male with schwannoma
125 mg/day, TIW
Diarrhea (grade 3)
59-year-old female with breast cancer
125 mg/day, TIW
Vomiting (grade 3)
C1D1—Patient had CTCAE grade 1 vomiting post dose despite predose of palonosetron. C1D3—Patient received palonosetron and dexamethasone as premedication and still suffered post dose CTCAE grade 1 vomiting and CTCAE grade 2 diarrhea. C1D6—Patient was premedicated with prochlorperazine and dexamethasone and experienced CTCAE grade 3 vomiting and diarrhea ∼5 h after dose. Vomiting was treated with prochlorperazine, and diarrhea was treated with lomotil. Both diarrhea and vomiting were considered as a DLT. C1D8—Patient discontinued study medication due to the adverse events suffered. C1—In the fourth week patient had diarrhea CTCAE grade 3, ∼4 h after dosing of BGT226. Despite optimal treatment of loperamide, patient had seven episodes of diarrhea (CTCAE grade 3) on two treatment days. This event was assessed as DLT, and the dose was reduced to 80 mg. Following dose reduction, the patient continued to experience CTCAE grade 1 diarrhea despite optimal treatment with loperamide. C1D8—Patient experienced CTCAE grade 1 vomiting post dose that resolved with palonosetron hydrochloride and lorazepam treatment. C1D15—Patient was premedicated with prochlorperazine. Post dose patient developed nausea/vomiting (CTCAE grade 3) and diarrhea (CTCAE grade 2). Patient was treated with lorazepam, prochlorperazine, lomotil, and prednisone. Diarrhea was not maximally treated. Vomiting was assessed as DLT, as it did not resolve with standard antiemetic treatment.
C, cycle; CTCAE, Common Terminology Criteria for Adverse Events; D, day; TIW, three times weekly.
Table 3. Patient exposure to BGT226, dose modifications, and disposition BGT226 dose cohort (mg/day) 2.5 (N = 5) 5 (N = 4) 10 (N = 4) 20 (N = 3) 40 (N = 8) 80 (N = 9) 100 (N = 13) 125 (N = 11) All (N = 57) Exposure, n (%) ≤ 4 weeks 2 (40) > 4 to ≤8 weeks 2 (40) > 8 to ≤12 weeks 0 > 12 weeks 1 (20) Median exposure, weeks 4.1 Minimum 1.1 Maximum 37.7 Patients with one or more dose reduction, n (%) 0 Due to AE 0 Patients with one or more dose delay due to AE, n (%) 2 (40) Treatment discontinued, n (%) 5 (100) Primary reason for treatment discontinuation, n (%) AE 0 Disease progression 5 (100) Consent withdrawal 0
1 (25) 1 (25) 1 (25) 1 (25) 8.0 3.7 31.7 0 0 0 4 (100)
1 (25) 2 (50) 0 1 (25) 5.5 2.4 19.7 0 0 0 4 (100)
0 0 4 (100) 2 (50) 0 2 (50)
1 (33) 2 (67) 0 0 4.1 3.9 7.9 0 0 0 3 (100)
1 (13) 4 (50) 2 (25) 1 (13) 7.7 0.9 15.9 0 0 1 (13) 8 (100)
1 (11) 4 (44) 0 4 (44) 7.7 3.4 31.6 1 (11) 0 3 (33) 9 (100)
3 (23) 4 (31) 3 (23) 3 (23) 7.9 3.7 48.1 1 (8) 0 6 (46) 13 (100)
3 (27) 5 (45) 2 (18) 1 (9) 7.6 0.9 12.3 4 (36) 4 (36) 7 (64) 11 (100)
13 (23) 24 (42) 8 (14) 12 (21) 7.6 0.9 48.1 6 (11) 4 (7) 19 (33) 57 (100)
0 3 (100) 0
0 8 (100) 0
1 (11) 8 (89) 0
0 13 (100) 0
3 (27) 8 (73) 0
4 (7) 51 (90) 2 (4)
Data cut-off 10 February 2010. AE, adverse event.
high and ranged (CV%) from 50% to 100% and from 30% to 200%, respectively, for most cohorts. Systemic exposure to BGT226 increased with increasing dose (Figure 1). It was, however, generally lower than expected based on preclinical data, and this finding was also supported by lower than
| Markman et al.
expected Cmax values, particularly at the 100 mg/day dose (Table 5). The median terminal elimination half-life ranged from 6 to 9 h after the first dose. No significant drug accumulation in plasma was reported following repeat administration.
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Patient
original articles
Annals of Oncology Table 4. Incidence of adverse events by BGT226 treatment cohort
BGT226 dose cohorts (mg/day) 2.5 (N = 5) 5 (N = 4) 10 (N = 4) 20 (N = 3) 40 (N = 8) 80 (N = 9) 100 (N = 13) 125 (N = 11) All (N = 57) All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 12 (92) 12 (92) 10 (77) 8 (62) 6 (46) 2 (15) 4 (31)
0 0 0 1 (8) 1 (8) 1 (8) 0
10 (91) 10 (91) 9 (82) 3 (27) 3 (27) 3 (27) 2 (18)
2 (18) 5 (46) 2 (18) 1 (9) 0 0 0
45 (79) 39 (68) 33 (58) 19 (33) 19 (33) 14 (25) 12 (21)
2 (4) 7 (12) 2 (4) 3 (5) 1 (2) 4 (7) 1 (2)
12 (92) 10 (77) 10 (77) 4 (31) 5 (39) 2 (15) 0
0 0 0 0 1 (8) 1 (8) 0
9 (82) 10 (91) 8 (73) 1 (9) 0 1 (9) 1 (9)
2 (18) 5 (46) 2 (18) 0 0 0 0
39 (68) 35 (61) 28 (49) 11 (19) 9 (16) 6 (11) 5 (9)
2 (4) 7 (12) 2 (4) 0 1 (2) 1 (2) 0
Gr, grade.
antitumor activity Stable disease (SD), defined as at least one SD assessment after start of treatment, was the best overall response achieved with BGT226 treatment. Response outcome as per investigator review is available for 51 patients; for 6 patients, there was either no documentation of progression or one or more target lesions were not assessed or were assessed using a different method than baseline. Seventeen of 57 patients (30%) achieved SD as best response. Tumor shrinkage, not qualifying for partial response, was observed in three of 57 patients (two with SD, one unknown best response). Nine patients (16%) had a progression-free period of four or more cycles. Thirty-four patients (60%) had progressive disease. Thirty patients (53%) achieved stable metabolic disease (SMD) as best metabolic response by central review of FDG– PET imaging. Metabolic progressive disease was reported in 10 patients (18%), and in the remaining patients, the percentage change from baseline in sSUVmax could not be assessed. Baseline and cycle 1 day 28 FDG–PET assessments were available for 38 patients (Figure 2). There was no apparent correlation between metabolic response by FDG–PET and standard radiological outcome measures determined by computed tomography (CT).
PD analysis Paired skin biopsies were available for 49 patients; 22 patients demonstrated > 25% decrease in pS6 levels at cycle 1, day 8 and day 28 compared with baseline (Figure 3A). However, consistent reduction in pS6 was present in relatively few cases, mostly in the 125-mg cohort, in which four of eight patients achieved reductions in pS6 (range 9% − 64% decrease) at both posttreatment time points; reduction was more pronounced
Volume 23 | No. 9 | September 2012
(range 37% − 64% decrease) after prolonged exposure to BGT226. A limited number of paired tumor biopsies were available (2.5-mg cohort, N = 3; 10-, 20-, and 40-mg cohorts, N = 1 each). Reductions in pS6 and pAkt levels, indicating inhibition of the PI3K pathway, were observed in tumor biopsies from the 2.5-mg and 20-mg cohorts, one of which (at 20 mg) also demonstrated a change in cellular proliferation (Figure 3B, C and D). Approximately half of the patients showed an increase in circulating cleaved and intact CK18, as measured by M30 (indicating epithelial cell apoptosis) and M65 (any type of cell death). Increases in total CK18 levels (M65) occurred mostly at BGT226 doses of 80 to 125 mg (data not shown). Changes in the circulating angiogenesis markers VEGF, sVEGFR1/2, PIGF, and bFGF, although evident in some patients, did not correlate with BGT226 dose levels (data not shown).
discussion In this first-in-human study, the MTD of the oral PI3K/mTOR inhibitor BGT226 in patients with advanced solid tumors was 125 mg/day, TIW; the recommended dose was defined as 100 mg administered once a day, TIW. The majority of patients (88%) experienced AEs potentially related to BGT226, with gastrointestinal effects and fatigue being the most frequent, consistent with preclinical observations. No grade >1 abnormalities in FPG were observed, in contrast to the inhibition of glucose metabolism and hyperglycemia associated with PI3K inhibition in BGT226 preclinical studies. The best overall response achieved with BGT226 was SD in 17 (30%) of patients. Tumor shrinkage not qualifying for partial response was observed in 3 of 57 patients (two with SD,
doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Most common adverse events, occurring in > 15% of patients, regardless of study drug relationship, n (%) Nausea 3 (60) 0 2 (50) 0 1 (25) 0 3 (100) 0 6 (75) 0 8 (89) 0 Diarrhea 1 (20) 0 1 (25) 0 1 (25) 0 2 (67) 0 4 (50) 0 8 (89) 2 (22) Vomiting 2 (40) 0 0 0 2 (50) 0 1 (33) 0 4 (50) 0 5 (56) 0 Decreased appetite 1 (20) 1 (20) 1 (25) 0 2 (50) 0 2 (67) 0 1 (13) 0 1 (11) 0 Fatigue 1 (20) 0 3 (75) 0 0 0 0 0 1 (13) 0 5 (56) 0 Anemia 2 (40) 2 (40) 1 (25) 0 1 (25) 1 (25) 1 (33) 0 2 (25) 0 2 (22) 0 Abdominal pain 1 (20) 1 (20) 1 (25) 0 1 (25) 0 1 (33) 0 1 (13) 0 1 (11) 0 Adverse events suspected to be related to study drug occurring in > 2 (4%) of patients, n (%) Nausea 1 (20) 0 1 (25) 0 1 (25) 0 2 (67) 0 5 (63) 0 8 (89) 0 Diarrhea 0 0 1 (25) 0 1 (25) 0 2 (67) 0 3 (38) 0 8 (89) 2 (22) Vomiting 0 0 0 0 1 (25) 0 1 (33) 0 3 (38) 0 5 (56) 0 Fatigue 0 0 0 0 0 0 0 0 1 (13) 0 5 (56) 0 Decreased appetite 0 0 0 0 2 (50) 0 1 (33) 0 0 0 1 (11) 0 Anemia 0 0 0 0 0 0 0 0 1 (13) 0 2 (22) 0 Asthenia 1 (20) 0 1 (25) 0 0 0 1 (33) 0 0 0 1 (11) 0
original articles
Annals of Oncology
Table 5. Summary of BGT226 pharmacokinetic parameters by cohort PK parameter (unit)
2.5 mg (N = 5)
AUC(0–48) (h*ng/ml) Cmax (ng/ml) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h) AUC(0–48) (h*ng/ml) Cmax (ng/ml) t1/2 (h) Tmax (h)
5 mg (N = 4)
10 mg (N = 4)
20 mg (N = 3)
40 mg (N = 8)
80 mg (N = 9)
100 mg (N = 13)
125 mg (N = 11)
Mean Mean (SD) Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean Mean (SD) Median Median Mean (SD) Mean (SD) Median Median Mean (SD) Mean (SD) Median Median
Cycle 1, day 1
Cycle 1, day 8
Cycle 2, day 1
— 0.18 (0.117) 1.53 1.73 (0.166) 0.42 (0.162) 9.22 1.09 4.45 (3.501) 0.81 (0.863) 6.31 2.55 9.79 (3.745) 1.47 (0.460) 8.35 2.00 11.67 (3.944) 1.47 (0.764) 7.76 3.00 32.86 (20.071) 2.50 (0.938) 7.61 3.00 26.97 (17.752) 2.45 (1.869) 8.61 3.00 78.75 (50.548) 5.21 (2.571) 8.62 4.00
0.49 0.11 (0.063) 0.97 2.29 (0.657) 0.31 (0.146) 17.74 1.99 5.37 (4.623) 0.78 (0.641) 8.79 2.25 14.41 (8.639) 1.86 (1.018) 7.83 3.12 12.12 (4.744) 0.92 (0.533) 8.86 2.48 32.58 (20.827) 2.70 (1.464) 8.03 3.00 32.11 (29.674) 2.59 (2.330) 9.84 3.00 86.66 (80.831) 4.40 (2.540) 7.09 3.95
— 0.13 (0.104) 2.50 1.49 (0.274) 0.34 (0.225) 13.80 0.97 3.62 (2.504) 0.38 (0.232) 9.98 2.99 15.35 (18.362) 1.67 (1.677) 7.64 4.61 12.07 (4.012) 1.47 (0.449) 11.45 2.26 30.52 (22.125) 3.07 (2.895) 9.16 3.00 26.41 (17.968) 2.43 (2.080) 8.18 2.00 68.89 (55.592) 4.14 (3.438) 11.10 4.54
Samples were collected predose and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 24, 32, and 48 h post dose. Not all patients in each cohort were assessable for all parameters for all time points. AUC, area under the plasma concentration versus time curve from 0 to 48 h (AUC0–48); Cmax, maximum plasma concentration; PK, pharmacokinetics; SD, standard deviation; Tmax, time to Cmax; t1/2, elimination half-life. a
one unknown). Nine patients (16%) remained progression free for ≥16 weeks. Thirty patients (53%) had SMD on FDG–PET evaluations; however, there was no apparent correlation between metabolic response and tumor shrinkage according to CT. It should be noted that the role of PI3K signaling in glucose uptake and metabolism [1] may confound interpretation of FDG–PET findings following treatment with BGT226, as detectable changes may reflect PI3K pathway downregulation more than antitumor activity. Changes in FDG–PET imaging with this class of agent may therefore have a PD component before indicating an antitumor effect [2]; however, further study is required. The PK profile of BGT226 was characterized by rapid absorption. The onset of vomiting on dosing days appeared to be correlated with Tmax/Cmax. Systemic exposure to BGT226 generally increased with increasing dose, and there was no significant accumulation of BGT226 following repeat dosing. In general, plasma levels of BGT226 were found to be 25- to
| Markman et al.
100-fold lower than expected based on preclinical data, which may have resulted in PK variability, suboptimal drug levels, and limited PD and antitumor activity. PD analyses suggested evidence of PI3K pathway inhibition following BGT226 treatment in selected patients. Overall, however, the PD effects of BGT226 were inconsistent. At BGT226 doses >80 mg, PI3K pathway inhibition was more evident in skin biopsies from a few patients, as determined by a reduction in pS6 (Ser240/244) by 37% − 64%. Available data suggest that BGT226 may also inhibit the PI3K pathway ( pAkt and pS6) in tumors; however, reductions of pS6/pAkt in tumors were not seen in all samples, and there were no available biopsies at doses ≥80 mg, making further interpretation difficult. These findings suggest that there was insufficient inhibition of the PI3K pathway in tumors by BGT226 to translate into significant antitumor effects with the explored dosing schedule. Consistently, no objective responses were observed with BGT226. Modeling based on PK data from this study and subsequent preclinical PK/PD analysis predicted
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
BGT226 dose cohorta
Annals of Oncology
original articles
that BGT226 doses >4000 mg/day would be required to achieve efficacious plasma exposure, greatly exceeding the human safety dose range. The four class I PI3K isoforms have overlapping nonredundant roles in cellular signal transduction [16]; however, preclinical data suggest context-specific functions. p110α is frequently mutated or amplified in human cancers, whereas overexpression of wild-type p110 β, δ, and γ isoforms can induce neoplastic transformation in cultured cells [3]. The PI3K inhibitors in clinical development achieve differing potencies against these isoforms, leading to speculation that the inhibitory profile of any given compound may have a significant impact on tolerability and activity. Preliminary clinical evidence suggests that this may be the case. Recent phase I study results of the novel dual PI3K/mTOR inhibitor BEZ235 in patients with advanced solid tumors showed that BEZ235 was well tolerated with a favorable safety profile and manageable gastrointestinal side-effects [17]. Notably, in the BEZ235 study, there were two partial responses in a patient with breast cancer and in a patient with lung cancer and Cowden syndrome [2]. In addition, recent results from a phase I study of the selective pan-class I PI3K inhibitor BKM120 also demonstrated a favorable safety and activity profile, including a partial response in a patient with breast cancer [18]. Preclinical data strongly support the use of PI3K inhibitors in tumors with PI3K pathway genetic alterations [9, 10, 19, 20]. Originally, it was intended in this study to investigate the relation between tumor mutational status and tumor response. However, as no objective responses were observed with BGT226, this comparison is obsolete and no conclusion to guide patient selection based on molecular tumor status can be drawn from this study. This may be due to insufficient
Volume 23 | No. 9 | September 2012
exposure, although the role of cellular signaling cross talk/ feedback (i.e. via the Ras pathway) cannot be excluded. It should be noted that patients both with and without identified tumor PI3K pathway alterations have demonstrated responses to other PI3K pathway inhibitors, including BEZ235 and BKM120 [17, 18]. BGT226 is one of the first compounds to be evaluated in the emerging PI3K inhibitor clinical development program. Although a small number of patients in our study derived some clinical benefit from BGT226, overall, the data suggest that it is not feasible to consistently achieve clinically significant antitumor effects with BGT226 at safely administered doses. Despite these findings, these data are valuable for ongoing and future studies of other PI3K-targeted inhibitors.
acknowledgements We thank the following investigators, Steven Isakoff (Massachussets General Hospital, Boston, USA), Sunil Sharma (Nevada Cancer Institute, Las Vegas, USA), Christophe LeTourneau (Princess Margaret Hospital, Toronto, Canada), and contributors Beatrix Staffler, Humphrey Gardner, Nina Huseinovic, Nassim Sleiman, Mikael Sacco, Hartmut Zehender, and Stefan de Buck (Novartis Pharma AG), whose efforts were essential toward the completion of this work. We would also like to thank the European Society of Medical Oncology (ESMO) for awarding BM with an ESMO fellowship grant that contributed to his involvement in this trial. The authors would also like to acknowledge Dina Marenstein, PhD, of Chameleon Communications International, who provided medical writing services with funding from Novartis Pharmaceuticals Corporation.
doi:10.1093/annonc/mds011 |
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 1. Plasma pharmacokinetic profile of BGT226 in patients with solid tumors. Mean plasma area under the curve (AUCinf ) values are shown for each dose cohort for cycle 1, day 1 and day 8, and cycle 2, day 1 time points.
original articles
Annals of Oncology
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 2. (A) Best percentage change from baseline in sum of longest diameters for 42 patients, best overall response as per central review by indication. (B) Tumor metabolic response to BGT226 therapy by [18F]-fluorodeoxyglucose–positron emission tomography (FDG–PET). Best percentage change from baseline to cycle 1, day 28 in the sum of maximum standardized uptake values (sSUVmax) for 38 patients with baseline and cycle 1, day 28 assessments.
| Markman et al.
Volume 23 | No. 9 | September 2012
Annals of Oncology
original articles
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
Figure 3. Pharmacodynamic effects of BGT226 treatment in skin and tumor biopsies. (A) Effect of BGT226 treatment on pS6 levels in skin biopsies. Data are presented as percent change from baseline of the H-score of the whole epidermis for evaluable posttreatment (day 8 and day 28 of cycle 1) samples by dose cohort. Effect of BGT226 treatment on (B) pS6, (C) pAkt, and (D) Ki-67 levels in tumor biopsies from a patient with colon cancer treated with 20 mg BGT226 ( percent change from baseline to cycle 2, day 1 is shown).
Volume 23 | No. 9 | September 2012
doi:10.1093/annonc/mds011 |
original articles funding Novartis Pharmaceuticals Corporation.
disclosure MMC, SS, DS, OA, and WH are employees of Novartis Pharma AG, the developer of BGT226 and the study sponsor. IK receives clinical trial funding from Novartis Pharma AG. JB has received consulting honoraria from Novartis. BM, JT, GIS, LS, LCC, and MM have no conflict of interest to declare.
| Markman et al.
10. Serra V, Markman B, Scaltriti M et al. NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Res 2008; 68: 8022–8030. 11. Maira SM, Voliva C, Garcia-Echeverria C. Class IA phosphatidylinositol 3-kinase: from their biologic implication in human cancers to drug discovery. Expert Opin Ther Targets 2008; 12: 223–238. 12. Maira SM, Stauffer F, Brueggen J et al. Identification and characterization of NVP-BEZ235, a new orally available dual phosphatidylinositol 3-kinase/ mammalian target of rapamycin inhibitor with potent in vivo antitumor activity. Mol Cancer Ther 2008; 7: 1851–1863. 13. Neuenschwander B, Branson M, Gsponer T. Critical aspects of the Bayesian approach to phase I cancer trials. Stat Med 2008; 27: 2420–2439. 14. Babb J, Rogatko A, Zacks S. Cancer phase I clinical trials: efficient dose escalation with overdose control. Stat Med 1998; 17: 1103–1120. 15. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 16. Engelman JA, Luo J, Cantley LC. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 2006; 7: 606–619. 17. Burris H, Rodon J, Sharma S et al. First-in-human phase I study of the oral PI3K inhibitor BEZ235 in patients with advanced solid tumors. Presented at the 2010 American Society of Clinical Oncology Annual Meeting (Abstr 3005). Chicago, IL June 4–8, 2010. 18. Bendell JC, Rodon J, Burris HA et al. Phase I, dose-escalation study of BKM120, an oral pan-class I PI3K inhibitor, in patients with advanced solid tumors. J Clin Oncol 2012; 30: 282–290. 19. Ihle NT, Lemos R, Jr, Wipf P et al. Mutations in the phosphatidylinositol-3-kinase pathway predict for antitumor activity of the inhibitor PX-866 whereas oncogenic Ras is a dominant predictor for resistance. Cancer Res 2009; 69: 143–150. 20. Engelman JA, Chen L, Tan X et al. Effective use of PI3K and MEK inhibitors to treat mutant Kras G12D and PIK3CA H1047R murine lung cancers. Nat Med 2008; 14: 1351–1356.
Volume 23 | No. 9 | September 2012
Downloaded from http://annonc.oxfordjournals.org/ at Fordham University on January 3, 2013
1. Katso R, Okkenhaug K, Ahmadi K et al. Cellular function of phosphoinositide 3kinases: implications for development, homeostasis, and cancer. Annu Rev Cell Dev Biol 2001; 17: 615–675. 2. Engelman JA. Targeting PI3K signalling in cancer: opportunities, challenges and limitations. Nat Rev Cancer 2009; 9: 550–562. 3. Courtney KD, Corcoran RB, Engelman JA. The PI3K pathway as drug target in human cancer. J Clin Oncol 2010; 28: 1075–1083. 4. Liu P, Cheng H, Roberts TM et al. Targeting the phosphoinositide 3-kinase pathway in cancer. Nat Rev Drug Discov 2009; 8: 627–644. 5. Vogt PK, Kang S, Elsliger MA et al. Cancer-specific mutations in phosphatidylinositol 3-kinase. Trends Biochem Sci 2007; 32: 342–349. 6. Brognard J, Clark AS, Ni Y et al. Akt/protein kinase B is constitutively active in non-small cell lung cancer cells and promotes cellular survival and resistance to chemotherapy and radiation. Cancer Res 2001; 61: 3986–3997. 7. Berns K, Horlings HM, Hennessy BT et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell 2007; 12: 395–402. 8. Hu L, Hofmann J, Lu Y et al. Inhibition of phosphatidylinositol 3’-kinase increases efficacy of paclitaxel in in vitro and in vivo ovarian cancer models. Cancer Res 2002; 62: 1087–1092. 9. Eichhorn PJ, Gili M, Scaltriti M et al. Phosphatidylinositol 3-kinase hyperactivation results in lapatinib resistance that is reversed by the mTOR/phosphatidylinositol 3kinase inhibitor NVP-BEZ235. Cancer Res 2008; 68: 9221–9230.
Annals of Oncology