A phase I dose-escalation study of the safety
and pharmacokinetics of a tablet formulation of voxtalisib,
a phosphoinositide 3-kinase inhibitor, in patients
with solid tumors

Janice M. Mehnert 1 & Gerald Edelman2 & Mark Stein1 & Heather Camisa1 &
Joanne Lager 3 & Jean-François Dedieu4 & Anne-Frédérique Ghuysen4 & Jyoti Sharma5 &
Li Liu5 & Patricia M. LoRusso6
Received: 15 March 2017 /Accepted: 5 April 2017
# Springer Science+Business Media New York 2017
Summary Background Voxtalisib, a PI3K/mTOR inhibitor,
has shown antitumor activity in capsule formulation in patients
with solid tumors. This Phase I study assessed safety and phar￾macokinetics of voxtalisib administered as immediate-release
tablets in patients with solid tumors (NCT01596270). Methods
A B3+3^ dose escalation design was used. Adverse events
(AEs), pharmacokinetics (PK), food effect and tumor response
were evaluated. Results Thirty-two patients received voxtalisib
doses ranging from 50 mg to 70 mg once daily (QD) and 17
patients received voxtalisib doses ranging from 30 mg to 50 mg
twice daily (BID), for two 28-day cycles. Dose-limiting toxic￾ities (DLTs) were Grade 3 fatigue (two patients at 70 mg QD,
one patient at 40 mg BID) and Grade 3 rash (two patients at
50 mg BID). The maximum tolerated dose (MTD) was 60 mg
for QD and 40 mg for BID regimens. Common treatment￾emergent AEs were diarrhea (41%), nausea (37%) and fatigue
(33%). Voxtalisib appeared to follow linear PK, with a general
increase in plasma exposure with dose and no significant accu￾mulation. Administration with food caused a slight decrease in
exposure; however, given the high variability observed in the
exposure parameters, this should be interpreted with caution.
Best response was stable disease in 29% and 50% of patients
(QD and BID regimens, respectively). Conclusions The safety
profile of voxtalisib tablets at the MTD in patients with solid
tumors was consistent with that observed with voxtalisib cap￾sules. Given the limited activity observed across multiple clin￾ical trials, no further trials of voxtalisib are planned.
Keywords PI3K . Voxtalisib . Tablets . Pharmacokinetics .
Solid tumors
Introduction
The phosphoinositide 3-kinase (PI3K) signaling pathway is
essential for cell growth and metabolism. PI3K activates the
protein serine-threonine kinase AKT, which results in the
stimulation of protein translation, cell proliferation, and inhi￾bition of apoptosis [1, 2]. AKT activates the mammalian target
of rapamycin (mTOR), an important effector of cell growth
and survival. Hyperactivation of the PI3K/AKT/mTOR path￾way has been implicated in the pathogenesis of many solid
tumors and hematologic malignancies [3–7]. Most of the on￾cogenic mutations occur in the p110α subunit of class IA
PI3Ks. Evidence suggests that treatment with isoform￾specific or pan-PI3K inhibitors may reduce tumor growth
and sensitize cancer cells to other therapies, such as chemo￾therapy or mitogen-activated protein kinase inhibitors [8, 9].
Inhibitors of both PI3K and mTOR have been shown to trig￾ger a more complete pathway inhibition, possibly through
additional inhibition of mTOR-dependent negative feedback
mechanisms [1, 2, 10, 11].
* Janice M. Mehnert
[email protected]
1 Rutgers Cancer Institute of New Jersey, 195 Little Albany Street,
New Brunswick, NJ 08901, USA
2 Medical and Surgical Clinic of Irving, Irving, TX, USA
3 Sanofi, Cambridge, MA, USA
4 Sanofi, Chilly-Mazarin, France
5 Sanofi, Bridgewater, NJ, USA
6 Yale University, New Haven, CT, USA
Invest New Drugs
DOI 10.1007/s10637-017-0467-7
Voxtalisib (SAR245409, XL765) is a specific, potent, and
reversible pan-Class I PI3K inhibitor and mTOR inhibitor
(both mTORC1 and mTORC2 complexes) [12, 13].
Voxtalisib has been shown to exert antitumor activity in human
xenografts by inhibiting tumor angiogenesis and inducing ap￾optosis in cancer cells [13, 14]. A Phase I study assessed the
safety and antitumor activity of voxtalisib in capsule formula￾tion in patients with solid tumors or lymphomas [15, 16].
Voxtalisib demonstrated reduced PI3K and mTORC1/
mTORC2 pathway activity in hair sheath cells, skin, and tu￾mor; best response was stable disease (SD) in 24 out of 50
(48%) evaluable patients with advanced solid tumors [16].
Preliminary clinical activity was also observed in patients with
lymphoma, with one complete response and two partial re￾sponses observed in 12 evaluable patients [15]. The safety
profile of voxtalisib was similar to those observed with other
PI3K inhibitors, with gastrointestinal toxicities, rash, and in￾creased liver enzymes as most common adverse events (AEs)
[16–21]. The maximum-tolerated dose (MTD) of voxtalisib
administered in capsule formulation was established at 50 mg
twice daily (BID) for patients with lymphomas [15] and at
50 mg BID or 90 mg once daily (QD) for patients with solid
tumors [16].
This Phase I study (NCT01596270) was conducted to as￾sess the safety, pharmacokinetics (PK), and antitumor activity
of an immediate-release, film-coated tablet formulation of
voxtalisib in patients with solid tumors or lymphoma. The
tablet was introduced as a dosage form more suitable for com￾mercialization than the capsule formulation of the drug.
Methods
Patient population
Eligible patients were aged ≥18 years and had either a histolog￾ically confirmed solid tumor that was metastatic or
unresectable, or relapsed or refractory lymphoma. Standard
therapies were no longer effective, and no therapeutic alterna￾tive existed or no therapy was known to prolong survival.
Patients were required to have an Eastern Cooperative
Oncology Group performance status (ECOG PS) ≤1, and ade￾quate organ and marrow function, including fasting plasma
glucose <160 mg/dL. Patients were excluded if they underwent
prior treatment with a small-molecule kinase inhibitor or prior
radiation therapy within 2 weeks before the first dose of study
treatment, or any other investigational therapy within 4 weeks
before the first dose of study treatment.
Study design
This was a Phase I, multicenter, open-label, non-randomized,
dose-escalation study with a cohort expansion phase in patients
with solid tumors or lymphoma (NCT01596270). Patients were
to receive voxtalisib QD or BID during two 28-day treatment
cycles. Treatment was to be discontinued in the case of disease
progression or unacceptable toxicity. The initial dose level was
50 mg QD; additional cohorts of 3 or 6 patients were to be
treated at successively higher doses in the QD treatment sched￾ule and in the BID treatment schedule in parallel. To determine
food interaction, patients received a single dose of 50 mg
voxtalisib in fasted condition and a single dose of 50 mg
voxtalisib in fed condition (standardized moderate-fat break￾fast), in a sequence determined as per the randomization list
with a minimum 48-h, maximum 4-day washout period be￾tween the two administrations. Additional patients (up to a total
of 15 patients evaluated in each treatment schedule) were to be
included in expansion cohorts at the QD and BID dose levels
determined to provide plasma exposure comparable to that at
the capsule MTD.
Patients eligible for treatment continuation after Cycle 2
were offered the opportunity to enroll in the treatment￾extension study NCT01587040. Patients not eligible for treat￾ment continuation after Cycle 2 were followed up for safety
for a minimum of 30 days (± 3 days) after last study drug
administration. The primary objective was to evaluate the
safety and tolerability of voxtalisib administered as a tablet
formulation QD or BID in patients with solid tumors or lym￾phoma. Secondary and exploratory objectives were to explore
PK, food interaction, and antitumor activity of voxtalisib.
This study was conducted in compliance with the recom￾mendations of the Helsinki Declaration, all relevant interna￾tional guidelines, and national laws and regulations. Informed
consent was obtained from all patients.
Dose escalation and dose-limiting toxicities
A B3+3^ dose-escalation design was employed. The dose￾limiting toxicity (DLT)-evaluable population consisted of all
included patients who had complete assessments for DLT
evaluation and received at least 80% of the doses of study
treatment during Cycle 1, or received partial doses of study
treatment but developed DLTs during Cycle 1.
A DLT was defined as any of the following occurring dur￾ing the first cycle of study treatment: any Grade ≥ 3 non￾hematologic toxicity (except diarrhea, nausea, or vomiting,
unless Grade 4 or lasting >2 days despite receiving optimal
prophylaxis and/or treatment); any toxicity resulting in a treat￾ment delay >2 weeks; Grade 4 neutropenia lasting ≥4 days;
febrile neutropenia defined as Grade ≥ 3 neutropenia associ￾ated with a single temperature > 38.3 °C or a sustained
temperature ≥ 38 °C for more than 1 h; neutropenic infection;
any other Grade 4 hematologic toxicity; or a treatment￾emergent AE (TEAE) that, in the opinion of the investigator,
was of potential clinical significance such that further dose￾escalation would expose patients to unacceptable risk. The
Invest New Drugs
dose escalation decision was based on study treatment-related
DLTs observed during Cycle 1 from at least 3 patients
in the DLT-evaluable population and on PK data.
Cumulative toxicities observed in subsequent cycle were
also to be considered for dose escalation and dose selection
decisions.
Safety assessments
Safety of voxtalisib was assessed by evaluation of TEAEs,
serious adverse events (SAEs), and DLTs graded with
National Cancer Institute Common Terminology Criteria for
Adverse Events (NCI CTCAE) (version 4.03); laboratory pa￾rameters; vital signs; electrocardiogram (ECG) parameters;
physical examination findings; ophthalmologic examinations
findings; and ECOG PS.
Pharmacokinetic assessments
Plasma concentrations of voxtalisib were measured on Cycle
1, Day 1 and Day 28 (pre-dose and at 0.5, 1, 1.5, 2, 3, 4, 6, 8
and 12 h for BID regimen and pre-dose; 0.5, 1, 1.5, 2, 3, 4, 6,
8, 12 and 24 h for QD regimen), and on Cycle 1, Day 2, Day 8
and Day 15 pre-dose using a validated liquid chromatography
coupled with tandem mass spectrometry (LC-MS/MS) meth￾od with a lower limit of quantification (LLOQ) of 1.00 ng/mL.
For food interaction investigation, plasma concentrations were
measured pre-dose and 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12 and 24 h
post-dose on Cycle 1, Day 1 and on Cycle 1, Day 3 in fasted
and fed conditions, respectively. Calculated PK parameters
included: maximum plasma concentration (Cmax); first time
to reach Cmax (tmax); area under the plasma concentration ver￾sus time curve calculated using the trapezoidal method from
time zero to the time of last measurable concentration
(AUClast) after single dose (Cycle 1, Day 1); area under the
plasma concentration versus time curve calculated using the
trapezoidal method from time zero to the end of the dosing
interval (AUCtau) on Cycle 1, Day 28; area under the plasma
concentration versus time curve extrapolated to infinity
(AUC); and plasma concentration observed just before treat￾ment administration during repeated dosing (Ctrough).
Voxtalisib PK parameters were summarized using descriptive
statistics. For the food interaction investigation only, the dif￾ference between food conditions was assessed on log￾transformed Cmax, AUClast, and AUC with a linear mixed
effects model. Estimate and 90% confidence interval (CI) for
the geometric means ratio between fed and fasted food condi￾tions were provided for Cmax, AUClast, and AUC. The PK
parameters were calculated using non-compartmental
methods with validated software (PKDMS [version 2.0] with
WinNonlin Professional, Pharsight [version 5.2.1]).
Efficacy assessments
Tumor assessment was performed using computed tomogra￾phy (CT) or magnetic resonance imaging (MRI) at investiga￾tor’s choice, at baseline, and at the end-of-treatment visit.
Tumor response was assessed by the investigator.
Table 1 Baseline characteristics
Voxtalisib QD
dosing regimen

ECOG PS Eastern Cooperative Oncology Group performance status
a
Other primary tumor sites refer to anal canal, Ewing’s sarcoma, extrahepatic bile duct, melanoma, oropharyn￾geal, pancreas, rectum/colon, renal cell, skin lower limb, soft tissue, thyroid, and unknown (one patient each)
b Other primary tumor sites refer to omentum, pancreatic, pelvis, testicular, and thymoma (one patient each)
Invest New Drugs
Results
Patient population
Forty-nine patients were enrolled and treated: three patients in
the 50 mg QD dose cohort, 12 patients in the 70 mg QD dose
cohort, 17 patients in the 60 mg QD dose cohort, four patients
in the 30 mg BID dose cohort, six patients in the 40 mg BID
dose cohort, and seven patients in the 50 mg BID dose cohort.
The median age of patients in the QD and BID cohorts was
56 years (range, 35–78) and 62 years (29–78), respectively
(Table 1). The most common tumor types at initial diagnosis
were lung cancer (25% in the QD cohorts, 18% in the BID
cohorts), colon cancer (19% in the QD cohorts, 24% in the
BID cohorts), and breast cancer (3.1% in the QD cohorts, 18%
in the BID cohorts). No patients with relapse or refractory
lymphoma were enrolled in the study. The median number
of prior anticancer therapies in the QD and BID cohorts were
3 (range, 1–12) and 4 (1–17), respectively.
The median duration of treatment in this trial in the QD
cohorts was 5.9 weeks (range, 0.9–8.1); median relative dose
intensity was 71%. The median duration of treatment in this
trial in the BID cohorts was 7.7 weeks (range, 0.9–8.1); the
median relative dose intensity was 83%. Fourteen patients
(44%) in the QD cohorts and 10 patients (59%) in the BID
cohorts completed the 8-week study treatment period. Early
study treatment discontinuation was reported for 18 patients
(out of 32) in the QD cohorts (including three due to
AE and 13 due to disease progression) and 7 patients (out of
17) in the BID cohorts (including one due to AE and five due
to disease progression).
Dose escalation and dose-limiting toxicities
For the QD dosing regimen, 60 mg QD was determined as the
MTD; the maximum administered dose (MAD) was 70 mg
QD. At the 60 mg dose level, additional patients were
included, up to a total of 15 DLT-evaluable patients.
Overall, at the QD MTD, none of the DLT-evaluable patients
experienced a DLT during Cycle 1. At the QD MAD, two out
of six (33%) DLT-evaluable patients experienced a DLT (se￾rious Grade 3 fatigue).
For the BID dosing regimen, 40 mg BID was determined as
the MTD; the MAD was 50 mg BID. At the BID MTD, one
out of six (17%) of DLT-evaluable patients experienced a DLT
(serious Grade 3 fatigue). At the BID MAD, two out of six
(33%) DLT-evaluable patients experienced a DLT (one was
non-serious Grade 3 maculopapular rash, the other was non￾serious Grade 3 macular rash. Both were resolved with corti￾costeroids. No additional patient was enrolled in the expan￾sion cohort at the BID MTD dose level since development of
voxtalisib was stopped.

BID twice daily, QD once daily, TEAE treatment emergent adverse event
Invest New Drugs
Safety
The most frequently reported TEAEs in the QD cohorts were
diarrhea (38%), nausea (38%), dyspnea (25%), and fatigue
(25%) (Table 2). The most frequently reported TEAEs in the
BID cohorts were diarrhea (47%), fatigue (47%), de￾creased appetite (41%), nausea (35%), cough (29%),
and vomiting (29%). The most frequently reported
Grade ≥ 3 TEAEs in the QD cohorts were disease progression
(12%), anemia (9%), fatigue (6%), dyspnea (6%), and
hyponatremia (6%). In the BID cohorts, the most frequently
reported Grade ≥ 3 TEAEs were disease progression (12%)
and hyponatremia (12%).
The most frequently reported treatment-related TEAEs in
the QD cohorts were nausea (34%), diarrhea (28%), decreased
appetite (13%), and fatigue (13%). The most frequently report￾ed treatment-related TEAEs in the BID cohorts were nausea
(29%), diarrhea (24%), fatigue (24%), and vomiting (24%).
Of the treatment-related Grade ≥ 3 TEAEs, only fatigue
occurred in more than one patient (two patients in the
QD cohorts).
The most frequently reported clinical chemistry abnormal￾ities during the on-treatment period in QD and BID cohorts,
respectively, were hyperglycemia (69% and 88%), hypoalbu￾minemia (56% and 65%), and alkaline phosphatase (ALP)
increased (53% and 53%). Grade 3 clinical chemistry abnor￾malities included hyperglycemia in two patients, hypoalbu￾minemia in two patients, ALP increased in two patients, and
ALT increased in one patient. One patient had ALT and AST
increases of >3 x upper limit of normal (ULN) and concurrent
total bilirubin values >2 x ULN with elevated ALP, which was
consistent with the presence of documented liver metastases.
The most frequently reported hematologic laboratory abnor￾malities during the on-treatment period in QD and BID
Table 3 Plasma pharmacokinetic parameters (mean ± SD (geometric mean) [%CV]) of voxtalisib on Cycle 1, Day 1 and Day 28

, ng•h/mL 920 ± NC 1530 ± 752 3390 ± 4310 1460 ± 199 3700 ± 6090 2190 ± 439
(920) [NC] (1410) [49.2] (1680) [127.1] (1450) [13.7] (2060) [164.3] (2160) [20.1]
AUC area under the concentration-time curve, AUC0-tau, AUClast, AUC from 0 h to the last measurable concentration, BID twice daily, Cmax maximum
concentration observed, h hour, NA not applicable, NC not calculated, QD once daily, t1/2z terminal half-life, tmax time to maximum plasma concentration

Invest New Drugs
cohorts, respectively, were anemia (81% and 100%), lympho￾penia (72% and 77%), leukopenia (34% and 41%), and throm￾bocytopenia (22% and 41%). The only Grade 4 hematologic
abnormality was lymphocyte count decreased in two patients
(one in each dose regimen).
The most frequently reported treatment-emergent SAEs
were disease progression (13%), fatigue (6%), and pyrexia
(6%) in the QD cohorts, and disease progression (12%) in
the BID cohorts. TEAEs leading to dose reduction were
hypophosphatemia and blood creatinine increased (one pa￾tient each) in the QD cohorts, and rash macular and ALT
increased (one patient each) in the BID cohorts. TEAEs lead￾ing to study treatment discontinuation were hematochezia,
hyperbilirubinemia (one patient each), ALT increased and
AST increased (both occurred in the same patient) in the QD
cohorts, and abdominal infection (one patient) in the BID
cohorts. A total of 11 deaths occurred during the study (seven
in the QD cohorts, four in the BID cohorts): seven due to
disease progression (including one due to an SAE of small
intestinal obstruction secondary to disease progression in the
BID cohorts); one due to an SAE of respiratory failure (QD
cohorts; SAE not related to study treatment); three during the
post-treatment follow-up period due to unknown reasons.
Pharmacokinetics
Summary of PK parameters of voxtalisib are shown in
Table 3. In general, plasma exposure of voxtalisib increased
with dose (Fig. 1) and a high variability (%CV) was observed
in the plasma exposures, with a %CVof up to 88.6% for Cmax
and up to 85.9% for AUClast on Day 1. Since the studied dose
levels are very close together (about 1.7-fold dose range), a
formal assessment of dose proportionality was not warranted.
No significant accumulation was observed following repeated
dosing for 28 days, as expected due to the short half-life of
voxtalisib (about 3–4 h). However, the accumulation esti￾mates with BID regimen were slightly greater (up to 1.99) as
compared to the QD regimen (up to 1.93). A visual examina￾tion of mean trough profiles suggested that steady state was
achieved on Day 15 (Fig. 2). Administration with food result￾ed in a slight decrease in plasma exposure (Table 4 and Fig. 3),

Day 1 Day 2 Day 8 Day 15 Day 28
Mean (SD) plasma voxtalisib concentration
(ng/ml)
Fig. 2 Mean plasma trough concentration of voxtalisib
Table 4 Food effect on voxtalisib plasma pharmacokinetic parameters
Mean ± SD (Geometric Mean)

AUC area under the concentration-time curve, AUC0–24 AUC from 0 h to
24 h, BID twice daily, Cmax maximum concentration observed, CV coef￾ficient of the variation, h hour, SD standard deviation, QD once daily, tmax
time to maximum plasma concentration
a
Median (min–max)
Invest New Drugs
however, given the %CVobserved in the exposure parameters
with and without food (Cmax and AUC), this result should be
interpreted with caution. A formal statistical analysis of food
effect provided the same conclusion, where the point estimates
of ratio of least square geometric means for Cmax and AUC
suggested that food decreases the plasma exposure (Table 5);
however, the very wide CIs around these point estimates war￾rant caution in interpreting this result.
Efficacy
Of 31 patients evaluable for antitumor activity in the QD co￾horts, nine patients (29%) had SD as best response. Of 16
patients evaluable for antitumor activity in the BID cohorts,
eight patients (50%) had SD as best response.
Discussion
This Phase I dose-escalation study evaluated the immediate
release, film-coated tablet formulation of voxtalisib in patients
with solid tumors. The safety profile of voxtalisib tablets at the
MTD in patients with solid tumors was generally similar to
that observed with voxtalisib capsules at the MTD [15, 16]; no
new safety issues were identified. The most frequently report￾ed TEAEs were diarrhea, nausea, and fatigue (41%, 37%, and
33% of all patients, respectively), consistent with the safety
profiles observed with other PI3K inhibitors [16–21]. Of note,
hyperglycemia, which often contributes to DLTs of pan-PI3K
inhibitors [22] was not a DLT in this study.
The MTD of the tablet formulation of voxtalisib was deter￾mined as 60 mg QD and 40 mg BID. In comparison, the MTD
of the capsule formulation was 90 mg QD or 50 mg BID [16].
Of note, doses are expressed in mg of voxtalisib free base for
tablets and mg of voxtalisib-hydrochloride salt for capsules
(with correction factor of 0.88), effectively meaning that a
90 mg capsule dose is equivalent to a 79.2 mg tablet dose.
Given the variability of PK parameters, the exposure achieved
with 60 mg QD tablet formulation on Cycle 1, Day 28 was
similar to that achieved with 90 mg QD capsule formulation
[16]. Voxtalisib appeared to follow linear PK, with a general
increase in plasma exposure with dose with both QD and BID
dosing regimens; given the high inter-individual variability,
small patient numbers, and the dose levels being very close
together, formal dose proportionality assessment was not war￾ranted. Moreover, MTD determination can have some uncer￾tainty in small size studies.
No significant accumulation of voxtalisib was observed
following repeated dosing for 28 days as expected due to the
short half-life of voxtalisib (about 3–4 h). However, the accu￾mulation estimates with BID regimen were slightly greater (up
to 1.99) than the QD regimen (up to 1.93). Administration
with food caused a slight decrease in voxtalisib exposure;
however, given the %CVobserved in the exposure parameters
with and without food (Cmax and AUC), this result should be
interpreted with caution.
Best response was SD in nine patients (29%) receiving QD
regimen and in 8 patients (50%) receiving BID regimen.
However, the response was evaluated by the investigator,
and no tumor measurement data was collected in this study.
According to the study design, disease assessment data were
not collected beyond two months. Observed efficacy was

Voxtalisib 50 mg Fasted
Voxtalisib 50 mg Fed
Fig. 3 Food effect on voxtalisib plasma pharmacokinetic parameters
Table 5 Statistical analysis of food effect on voxtalisib exposure
Factor Cmax, ng/ml AUClast, ng•hr/ml AUC0–24, ng•hr/ml AUC, ng•hr/ml

Pairwise comparison: Voxtalisib 50 mg fed (Test) vs. voxtalisib 50 mg fast (Ref)
Ratio of least square geometric means (%) 65 63 65 65
90% Confidence interval around ratio 42–101 38–106 39–107 39–107
Treatments = [Voxtalisib 50 mg Fast, Voxtalisib 50 mg Fed]
Invest New Drugs
consistent with previous studies of voxtalisib in patients with
solid tumors [16]. Of note, a recent study of voxtalisib in
patients with lymphomas reported one complete response
and two partial responses in 12 evaluable patients, suggesting
higher voxtalisib activity in patients with lymphomas [15].
In this study, patients were not selected based on tumor
molecular profiling, and the study did not utilize a biomarker￾stratified system to assess for possible differences in response
rates depending on the presence or absence of individual cancer
mutation status in the PI3K/mTOR pathway. Of note, one of the
patients with four-month long stable disease had a missense
mutation in the PI3K p110α subunit. Consistent with the results
observed in this trial, various PI3K inhibitors recently assessed
in early-phase trials demonstrated limited efficacy in solid tu￾mors in unselected patient populations [23, 24]. Toxicity and
suboptimal dose schedules or formulations represented a poten￾tial hurdle. Additionally, inhibiting one or several PI3K iso￾forms often elicits feedback response mediated by PTEN loss
and/or AKT or MAPK hyperactivation [24]. In future, it will be
important to identify biomarkers that would help select patients
who are likely to benefit from the particular treatment.
Narrowing down the patient population, treatment dose, sched￾ule optimization, and rational combinations with other agents
will help to identify the full potential of PI3K/mTOR inhibitors
in cancer treatment.
In summary, this Phase I dose-escalation study demonstrated
that dose regimens of 60 mg QD and 40 mg BID for the tablet
formulation of voxtalisib had a similar safety profile and similar
plasma exposure compared with the dose regimens of
90 mg QD and 50 mg BID with a capsule formulation evalu￾ated in previous trials. Based on the pharmacokinetic results,
the study concluded that the tablet and the capsule prepara￾tions were comparable. Given the limited activity observed
across multiple clinical trials, no further clinical trials of
voxtalisib are planned.
Acknowledgements This study was funded by Sanofi. The authors
received editorial support from Olga Ucar of MediTech Media, funded
by Sanofi.
Compliance with ethical standards
Conflict of interest JMM has participated in advisory boards for
Merck, Genentech, EMD Serono, performed a consultancy role for
Merck and Amgen, and received research funding from Polynone,
Merck, Amgen, Sanofi, EMD Serono, Novartis, AstraZeneca, and
Immunocore. GE, HC and PML have no conflict of interest to declare.
MS reports research funding to his institution from Oncoceutics, Merck
Sharp & Dohme, Abbvie, Janssen Oncology, Medivation/Astellas,
Astellas Pharma, Bavarian Nordic and Advaxis. JL and JS are employees
and shareholders of Sanofi. JFD, AFG and LL are remunerated em￾ployees and shareholders of Sanofi.
Funding This study was funded by Sanofi. The authors received edi￾torial support from Olga Ucar of MediTech Media, funded by Sanofi.
Ethical approval All procedures involving human participants per￾formed in this study were in accordance with the ethical standards of
the institutional and/or national research committee and with the
1964 Helsinki Declaration and its later amendments or compara￾ble ethical standards.
Informed consent Informed consent was obtained from all individual
participants included in the study.
References
1. van der Heijden MS, Bernards R (2010) Inhibition of the
PI3K pathway: hope we can believe in? Clin Cancer Res
16:3094–3099
2. Courtney KD, Corcoran RB, Engelman JA (2010) The PI3K path￾way as drug target in human cancer. J Clin Oncol 28:1075–1083
3. Cantley LC (2002) The phosphoinositide 3-kinase pathway.
Science 296:1655–1657
4. Guillermet-Guibert J, Bjorklof K, Salpekar A, Gonella C, Ramadani
F, Bilancio A, Meek S, Smith AJ, Okkenhaug K, Vanhaesebroeck B
(2008) The p110beta isoform of phosphoinositide 3-kinase signals
downstream of G protein-coupled receptors and is functionally re￾dundant with p110gamma. Proc Natl Acad Sci U S A 105:8292–
8297
5. Engelman JA, Luo J, Cantley LC (2006) The evolution of phos￾phatidylinositol 3-kinases as regulators of growth and metabolism.
Nat Rev Genet 7:606–619
6. Liu P, Cheng H, Roberts TM, Zhao JJ (2009) Targeting the
phosphoinositide 3-kinase pathway in cancer. Nat Rev Drug Discov
8:627–644
7. Zhao L, Vogt PK (2008) Class I PI3K in oncogenic cellular trans￾formation. Oncogene 27:5486–5496
8. Villanueva J, Vultur A, Lee JT, Somasundaram R, Fukunaga￾Kalabis M, Cipolla AK, Wubbenhorst B, Xu X, Gimotty PA, Kee
D, Santiago-Walker AE, Letrero R, D’Andrea K, Pushparajan A,
Hayden JE, Brown KD, Laquerre S, McArthur GA, Sosman JA,
Nathanson KL, Herlyn M (2010) Acquired resistance to BRAF
inhibitors mediated by a RAF kinase switch in melanoma can be
overcome by cotargeting MEK and IGF-1R/PI3K. Cancer Cell 18:
683–695
9. Greger JG, Eastman SD, Zhang V, Bleam MR, Hughes AM,
Smitheman KN, Dickerson SH, Laquerre SG, Liu L, Gilmer TM
(2012) Combinations of BRAF, MEK, and PI3K/mTOR inhibitors
overcome acquired resistance to the BRAF inhibitor GSK2118436
dabrafenib, mediated by NRAS or MEK mutations. Mol Cancer
Ther 11:909–920
10. O’Reilly KE, Rojo F, She QB, Solit D, Mills GB, Smith D, Lane H,
Hofmann F, Hicklin DJ, Ludwig DL, Baselga J, Rosen N (2006)
mTOR inhibition induces upstream receptor tyrosine kinase signal￾ing and activates Akt. Cancer Res 66:1500–1508
11. Muller A, Zang C, Chumduri C, Dorken B, Daniel PT, Scholz CW
(2013) Concurrent inhibition of PI3K and mTORC1/mTORC2
overcomes resistance to rapamycin induced apoptosis by down￾regulation of Mcl-1 in mantle cell lymphoma. Int J Cancer 133:
1813–1824
12. Markman B, Atzori F, Perez-Garcia J, Tabernero J, Baselga J (2010)
Status of PI3K inhibition and biomarker development in cancer
therapeutics. Ann Oncol 21:683–691
13. Markman B, Dienstmann R, Tabernero J (2010) Targeting the
PI3K/Akt/mTOR pathway–beyond rapalogs. Oncotarget 1:530–543
14. Prasad G, Sottero T, Yang X, Mueller S, James CD, Weiss WA,
Polley MY, Ozawa T, Berger MS, Aftab DT, Prados MD, Haas￾Kogan DA (2011) Inhibition of PI3K/mTOR pathways in
Invest New Drugs
glioblastoma and implications for combination therapy with temo￾zolomide. Neuro-Oncology 13:384–392
15. Papadopoulos KP, Egile C, Ruiz-Soto R, Jiang J, Shi W, Bentzien F,
Rasco D, Abrisqueta P, Vose JM, Tabernero J (2015) Efficacy, safe￾ty, pharmacokinetics and pharmacodynamics of SAR245409
(voxtalisib, XL765), an orally administered phosphoinositide 3-ki￾nase/mammalian target of rapamycin inhibitor: a phase 1 expansion
cohort in patients with relapsed or refractory lymphoma. Leuk
Lymphoma 56:1763–1770
16. Papadopoulos K, Tabernero J, Markman B, Patnaik A, Tolcher A,
Baselga J, Shi W, Egile C, Ruiz-Soto R, Laird AD, Miles D,
Lorusso PM (2014) Phase I safety, pharmacokinetic and pharmaco￾dynamic study of SAR245409 (XL765), a novel, orally adminis￾tered PI3K/mTOR inhibitor in patients with advanced solid tumors.
Clin Cancer Res 20:2445–2456
17. Markman B, Tabernero J, Krop I, Shapiro GI, Siu L, Chen LC, Mita
M, Melendez CM, Stutvoet S, Birle D, Anak O, Hackl W, Baselga J
(2012) Phase I safety, pharmacokinetic, and pharmacodynamic
study of the oral phosphatidylinositol-3-kinase and mTOR inhibitor
BGT226 in patients with advanced solid tumors. Ann Oncol 23:
2399–2408
18. Arkenau H, Fields Jones S, Kurkjian C, Infante JR, Pant S, Burris
HA, Moore KN, McMeekin DS, Greco FA, Ramsey S, Bendell JC
(2012) The PI3K/mTOR inhibitor BEZ235 given twice daily for the
treatment of patients (pts) with advanced solid tumors. J Clin Oncol
30:abstr.3097
19. Wagner AJ, Bendell JC, Dolly S, Morgan JA, Ware JA, Fredrickson
J, Mazina KE, Lauchle JO, Burris HA, De Bono JS (2011) A first￾in-human phase I study to evaluate GDC-0980, an oral PI3K/
mTOR inhibitor, administered QD in patients with advanced solid
tumors. J Clin Oncol 29:abstr.3020
20. Hong DS, Bowles DW, Falchook GS, Messersmith WA, George
GC, O’Bryant CL, Vo AC, Klucher K, Herbst RS, Eckhardt SG,
Peterson S, Hausman DF, Kurzrock R, Jimeno A (2012) A multi￾center phase I trial of PX-866, an oral irreversible phos￾phatidylinositol 3-kinase inhibitor, in patients with advanced solid
tumors. Clin Cancer Res 18:4173–4182
21. Shapiro GI, Rodon J, Bedell C, Kwak EL, Baselga J, Brana I,
Pandya SS, Scheffold C, Voxtalisib Laird AD, Nguyen LT, Xu Y, Egile C,
Edelman G (2014) Phase I safety, pharmacokinetic, and pharmaco￾dynamic study of SAR245408 (XL147), an oral pan-class I PI3K
inhibitor, in patients with advanced solid tumors. Clin Cancer Res
20:233–245
22. Akinleye A, Avvaru P, Furqan M, Song Y, Liu D (2013)
Phosphatidylinositol 3-kinase (PI3K) inhibitors as cancer therapeu￾tics. J Hematol Oncol 6:88
23. Fruman DA, Rommel C (2014) PI3K and cancer: lessons, chal￾lenges and opportunities. Nat Rev Drug Discov 13:140–156
24. LoRusso PM (2016) Inhibition of the PI3K/AKT/mTOR pathway