Phase I study of metformin and chloroquine in IDH1/2-mutated patients with intrahepatic cholangiocarcinoma or chondrosarcoma

Isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) are homodimeric
enzymes that reversibly convert isocitrate to *-ketoglutarate (*KG) in the
cytoplasm and mitochondria, respectively (1). Somatic heterozygous mutations in
IDH1/2 (IDH1/2MT) are observed in substantial percentages of various tumors
such as WHO grade II-III glioma (80%, ~1000 mutated patients annually (MPA)),
intrahepatic cholangiocarcinoma (IHCC; 20%, ~40 MPA) and chondrosarcoma (CS;
60%, ~140 MPA) (1). In addition, IDH1/2MT occur in varying percentages of acute
myeloid leukemia, acute lymphocytic leukemia, T-cell lymphoma and osteosarcoma.
Glioma comprise WHO grades II to IV and vary from low-grade astrocytoma and
oligodendroglioma, with median survivals > 5 years, to glioblastoma, with a
median survival of only 15 months under aggressive treatment with radiotherapy
and temozolomide (2). Glioma grow very diffusely and this renders surgery
ineffective in most cases, highlighting the dire need for novel therapeutic
modalities. Furthermore, the blood-brain barrier (BBB) prohibits the use of
most chemotherapeutics and the surrounding normal brain hampers aggressive
radiotherapy regimens in the context of radiotoxicity. Large proportions (~80%)
of secondary glioma and glioblastoma are affected by IDH1/2MT and these
mutations are considered inaugural genetic defects, and are thus present in a
large fraction, if not all, cancer cells (3). This makes IDH1/2MT an
interesting target for glioma treatment. Cholangiocarcinoma is resectable in
~30% of patients and irresectable in ~70% of patients (4). In resectable cases,
cholangiocarcinoma is currently not treated with adjuvant chemotherapy. In
irresectable cases, cholangiocarcinoma patients are offered palliative
treatment with the chemotherapy combination of cisplatin and gemcitabine,
although current median overall survival remains poor (11.7 months) (5). The
current standard therapy for CS is surgery, and there is no evidence for a
benefit of adjuvant radiotherapy or chemotherapy (6). Consequently, prognosis
of high-grade CS remains poor with a 1-year survival rate of less than 10%.
This underlines the dire need for novel therapies for these solid tumors.
Compared with healthy cells, cancer cells have an increased uptake of glutamine
to facilitate tricarboxylic acid (TCA) cycle influx under conditions of aerobic
glycolysis (also called the Warburg effect), and for the biosynthesis of
macromolecules needed for proliferation (7). After entering the cell, glutamine
is converted to glutamate by glutaminase (GLS, Fig. 1). Subsequently, glutamate
is converted to *KG by glutamate dehydrogenase (GDH) or glutamate
transaminases. This two-step process is also called glutaminolysis. All
IDH1/2MT induce a neo-enzymatic activity that leads to the accumulation of the
oncometabolite D-2-hydroxyglutarate (D-2HG) (8). The oncometabolite D-2HG is
produced from *KG, which is a product of either IDH1/2 wild-type (IDH1/2WT) or
glutaminolysis. In addition, all IDH1/2MT disable enzymatic IDH1/2WT function
via a dominant-negative effect (9), either through a structural mechanism (10)
or via a *pseudo-product* inhibition of accumulated D-2HG which mimics *KG
(unpublished observations). Since IDH1/2WT activity is impaired in IDH1/2MT
tumors, glutaminolysis is the most important source of *KG there. IDH1/2MT
tumors are addicted to the oncogenic effects of D-2HG, and are therefore
dependent on glutaminolysis for proper *KG supply (11). The glutamate-to-*KG
conversion by GDH, which is the final step of glutaminolysis, can be inhibited
by the anti-malaria drug chloroquine (12-15). In addition, IDH1MT glioma cells
have increased levels of autophagy, which may suggest IDH1/2MT induce
dependency of cells on autophagy to combat oxidative stress or to catabolize
macromolecules to generate ATP (16). Thus, the anti-cancer properties of
chloroquine may have a high selectivity for IDH1/2MT cells because it attacks
both the cells* dependency on glutaminolysis and autophagy. Chloroquine is a
cheap, readily available, widely used drug with a safety profile that is
favorable to that of most other chemotherapeutics. Figure 1. Cellular
carbohydrate metabolism. Hypoxic cells increase glutamine import and catabolism
via glutaminolysis. Both the accumulation of D-2HG and the disabling of
IDH1/2WT kinetics by IDH1/2MT induce metabolic stress in IDH1/2MT cancer cells.
This metabolic stress was amplified in vitro using pharmaceuticals that inhibit
complex I of the electron transport chain (ETC), such as the oral antidiabetic
biguanides metformin and phenformin (17). These inhibitors induced selective
growth rate reductions in several IDH1MT cells, but not IDH1/2WT cancer cells.
The lipophilic analogue of metformin, phenformin, preferentially slowed
inhibition of IDH1MT HCT116 colorectal carcinoma cells (40% reduction at 25
uM), compared with IDH1WT HCT116 colorectal carcinoma cells (6% reduction at 25
uM) (17). Even more than chloroquine, metformin is cheap, one of the most
widely used drugs in the world, and has a very good safety profile. Therefore,
it is tempting to investigate the effects of metformin and chloroquine in
patients with IDH1/2MT cancers. This study will be the first to determine the
recommended dose for repurposing metformin plus chloroquine, two cheap,
readily-available drugs with a very favorable safety prolife, in patients with
IDH1/2MT solid tumors. Thus far, there are no reports of clinical trials with a
combination schedule of metformin and chloroquine. Three rare cases of
hypoglycemia induced by chloroquine in non-diabetics have been described in the
literature (18-20), although the underlying mechanism is still unclear and
chloroquine does not induce decreases of serum glucose concentrations in
healthy individuals (21). However, these results imply that we need to closely
monitor blood glucose levels of patients on a combination therapy. Metformin is
not metabolized, but cleared via renal tubular secretion. This secretion is
inhibited by cationic compounds such as chloroquine (22), thus chloroquine may
increase serum metformin concentrations by inhibiting metformin clearance.
Conversely, a severe side effect of metformin is hepatotoxicity (frequency:
<0.01%) and chloroquine protected metformine-overdosed mice from these
hepatotoxic effects (23). In an in vitro study investigating the effect of
metformin and chloroquine on IDH1/2WT prostate cancer cells, the combination of
metformin and chloroquine at supraphysiological dose levels was well tolerated
(24), suggesting there is a therapeutic index of metformin and chloroquine in
IDH1/2MT cells over IDH1/2WT cells. The latter represent, in the context of
IDH1/2MT cancer patients, the healthy cells. The combination of metformin plus
chloroquine has never been investigated in patients. Therefore, this
investigation of metformin plus chloroquine in IDH1/2MT IHCC and CS patients is
primarily a phase Ib safety study which also investigates the effect of
metformin plus chloroquine on tumor metabolism, D-2HG levels and tumor
response. The golden standard of IDH1/2MT detection is sequencing of tumor
material. Non-invasively, the presence of IDH1/2MT in cholangiocarcinoma can be
easily detected via mass spectroscopy (MS) of D-2HG levels in serum or urine,
as is also done in acute myeloid leukemia (25). This method does not apply to
IDH1/2MT glioma, probably because D-2HG levels generated by IDH1/2MT glioma are
too low to detect in the serum or urine as D-2HG is poorly distributed over the
blood-brain barrier. Recently, an intriguing paper was published, showing that
the ratio of serum 2HG and urine 2HG (enantiomer unspecific) is predictive for
the presence of IDHMT glioma, owing to a higher urine 2HG concentration in
IDHWT glioma (26). These results are provocative and need to be confirmed in
additional studies. The presence of intratumoral D-2HG levels as a biomarker
for IDH1/2MT in glioma can be easily detected via magnetic resonance
spectroscopy (MRS), which is rendered as the golden standard for noninvasive
detection of IDH1/2MT for glioma. We hypothesize that MRS is also feasibile to
measure intratumoral D-2HG measurements in other solid tumors, such as
cholangiocarcinoma and CS. Because of the absence of a blood-brain barrier
around cholangiocarcinoma and CS, we also hypothesize that we can detect
IDH1/2MT in cholangiocarcinoma via D-2HG serum, urine or bile MS and IDH1/2MT
in CS via D-2HG serum or urine MS, analogous to IDH1/2MT AML patients and D-2HG
aciduria patients, where this is a routine procedure. Tumor material (e.g.
surgery performed before or after this clinical trial or a tumor biopsy
specifically for this clinical trial) will be sequenced using an IonTorrent
platform to validate our MS and MRS results. This will also yield sensitivity
and specificity parameters for these investigational diagnostic modalities.
Available tumor material from IDH1/2WT IHCC and CS patients, who will be
excluded from this clinical trial, will be sequenced for IDH1/2MT in the tumor
to constitute the control research population here. In summary, our hypothesis
is that we can use metformin and chloroquine as anti-cancer drug for IDH1/2MT
cholangiocarcinoma and CS patients and observe tumor response by measuring
tumor size, D-2HG serum/urine/bile and/or intratumoral levels. This will be
investigated in this phase Ib clinical trial.

Primary objective * To determine the maximum tolerated dose and recommended
dose of metformin plus chloroquine in patients with IDH1/2MT glioma, IHCC and
CS Secondary objectives * To describe the toxic effects of metformin plus
chloroquine in patients with IDH1/2MT glioma, IHCC and CS. * To describe the
pharmacokinetics of metformin plus chloroquine in patients with IDH1/2MT
glioma, IHCC and CS. * To validate whether the ratio of serum 2HG to urine 2HG
is predictive for IDH1/2MT glioma. * To provide evidence that IDH1/2MT status
can be determined by MRS-facilitated detection of intratumoral D-2HG levels or
next-generation sequencing (NGS) of circulating tumor DNA (ctDNA) in IHCC
patients * To provide evidence that IDH1/2MT status can be determined by
MRS-facilitated or MS-facilitated detection of intratumoral/serum/urine D-2HG
levels or NGS of ctDNA in CS patients * To provide evidence that IDH1/2MT
status can be determined by NGS of ctDNA in glioma patients. * To provide
preliminary evidence of metabolic activity of metformin plus chloroquine in
IDH1/2MT glioma, IHCC and CS by assessment of D-2HG accumulation in
serum/urine/bile (via MS) and the tumoral mass (via MRS). * To provide
preliminary evidence of (partial) IDH1/2MT tumor regression after treatment
with metformin plus chloroquine, as measured using a MRI/CT scan.

This is a nonrandomized, open-label, single-center phase Ib clinical trial.
Patients will be screened for inclusion criteria, including MS of
serum/urine/bile to detect whether or not their glioma, IHCC and CS carries a
D-2HG-generating mutation in IDH1 or IDH2. We aim to enroll 20 patients. The
patient will be requested to maintain a medication diary of each dose of
medication. The medication diary will be returned to clinic staff at the end of
each cycle. Starting Dose of study drugs The starting dose of metformin will be
500 mg po bid. This dose is based on an earlier phase II clinical trial in
pancreatic adenocarcinoma (27). Chloroquine will be added to metformin in week
2 of the study. Chloroquine doses will be fixed, i.e. not escalate during the
study. * For glioma patients and irresectable/metastasized IHCC and CS
patients, the chloroquine dose will be 200 mg qd. * For resectable IHCC and CS
patients, chloroquine will be given in a step-down dosing schedule. Because we
expect the study duration to be a few weeks (e.g. there is a usual waiting time
of 6-8 weeks from diagnosis until surgery for resectable cholangiocarcinoma
patients), this allows build-up of functional chloroquine serum concentrations
in a shorter time. o First 2 weeks of chloroquine administration (week 2 and 3
of study): 300 mg per os qd. o Subsequent weeks (week 4 of study and later):
200 mg per os qd. This dosing schedule is based on treatment schedules in
systematic lupus erythematodes (28). Dose-Escalation Schedule Metformin The
dose of metformin will be escalated in fixed increments according to the
dose-escalation scheme outlined in Table 1 of the protocol. Chloroquine The
dose of chloroquine will be fixed, i.e. not escalated. Methods and Endpoints
The rate of subject entry and escalation to the next dose level will depend
upon assessment of the safety profile of patients entered at the previous dose
level. Toxicity will be evaluated according to the NCI Common Terminology
Criteria for Adverse Events (CTCAE), Version 4.03 (see Appendix F). A minimum
of three patients will be entered on each dose level. All three will be
followed for one completed cycle of therapy (28 days) and subsequent enrolment
of new cohorts will be based on the toxicity assessment in that first cycle and
the documentation of any dose limiting toxicities (for definitions see below).
Intrapatient dose escalation is permitted as outlined below. Maximum
Administered Dose (MAD) If 0/3 patients exhibit dose limiting toxicity at this
dose level: * Dose escalation to the next dose level may begin in a new cohort
of patients * Patients enrolled on the previous dose level who are still
receiving therapy may now undergo intrapatient escalation to this dose level
provided they have experienced no drug related toxicity grade 2 or more.
Toxicity information at this new dose level for intrapatient dose-escalated
patients will be used for expansion cohorts (when 1/3 of a new cohort of
patients exhibit a dose limiting toxicity at the new dose level). If 1/3
patients exhibit dose limiting toxicity at this dose level: * Expand dose level
to a total of 6 patients. * If no further DLT events seen, dose escalation to
the next dose level may begin in a new cohort of patients and patients enrolled
on the previous dose level who are still receiving therapy may now undergo
intrapatient escalation to this dose level provided they have experienced no
drug related toxicity grade 2 or more. * If 1 or more further DLT events are
seen (i.e. 2 or more of 6 patients), this dose level will be considered the
maximum administered dose (MAD). If 2/3 patients exhibit dose limiting toxicity
* This dose level will be considered the maximum administered dose (MAD). * If
this toxicity occurs at level 1 (starting level), dose de-escalation to level
-1 will occur. Before opening the next higher dose level all toxic effects at
the preceding dose level will be reviewed and expansion or escalation will be
undertaken as appropriate. Conference calls between investigators will be
organized if required. Dose Limiting Toxicity (DLT) Toxicity will be graded
using CTCAE version 4.03 (see Appendix F). Any dose limiting toxicity must be a
toxicity that is considered related to study drug. Dose limiting toxicity is
defined as follows: Hematologic * Absolute granulocyte count (AGC) < 0.5 x
109/l. * Febrile Neutropenia (ANC < 1.0 x 109/L, fever > 38.5oC). * Platelets <
25 x 109/l. * Bleeding felt to be due to thrombocytopenia. Non-Hematologic: *
Diarrhea > Grade 3 despite optimal loperamide use. * Rash > Grade 3 or grade 2
is medically concerning or unacceptable to the patient. * Other grade 3 effects
thought to be treatment related. * Missing >7 days of treatment for toxicity
reasons. When a patient experiences a DLT, he/she can choose to a) withdraw
from the study, or b) go into intrapatient de-escalation by receiving metformin
at one dose level lower than the dose level that provoked the DLT. Recommended
Phase II Dose As described above the MAD is that dose in which 2/3 or 2/6
patients experience dose limiting toxicity. Normally one dose level below that
dose will be considered the recommended phase II dose. If the MAD is seen at
the starting dose level, then dose level *-1* will be the recommended dose,
providing that we observe no DLTs at this dose level. If clinically
appropriate, intermediate dose levels may be studied to assure that the
recommended dose is the highest tolerable. Further, if pharmacokinetic data
suggests that saturable absorption of drug is occurring on a b.i.d. oral
administration level, further dose splitting to t.i.d. or q.i.d. schedules may
be considered if dose limiting toxicity has not been seen. Up to a total of 6
patients may be treated at the recommended dose to assure information on the
safety profile when that dose is complete. Patient Replacement Three patients
within a dose level must be observed for one cycle (28 days) before accrual to
the next higher dose level may begin. If a patient is withdrawn from the study
prior to completing 22 days of therapy without experiencing a DLT prior to
withdrawal, an additional patient may be added to that dose level. Patients
missing 7 or more doses due to toxicity will not be replaced since these
patients will be considered to have experienced a dose limiting toxicity. In
the remainder of this chapter, we will separately discuss (4.1) patients with
resectable IHCC patients and resectable CS patients and (4.2) patients with
glioma or locally advanced (irresectable) or metastasized IHCC or CS patients.
Resectable IHCC and resectable CS patients The present study will be conducted
during the waiting period until surgery. By approval of the patient, D-2HG
concentration can already be measured during the patient*s consideration of
trial participation in order to prolong the study duration in which patients
can be treated. For this purpose, the patient separately gives informed
consent. The end of study is defined as when a patient chooses to withdraw from
the study (with or without experiencing a DLT), or 2 days before surgery.
Glioma patients, irresectable/metastasized IHCC patients and
irresectable/metastasized CS patients The study will not be conducted
simultaneously with radiotherapy and/or chemotherapy treatment. The end of
study is defined as when a patient chooses to withdraw from the study (with or
without experiencing a DLT), or when tumor progression occurs.

N/A

Metformin Based on human studies, major adverse effects of metformin which may
limit dose are gastro-intestinal side effects (e.g. nausea, diarrhoea). Rarely,
hypoglycaemia occurs. Rare, severe side-effects are lactic acidosis (frequency
< 0.01%) and hepatotoxicity (frequency 0.01%). Lactic acidosis is a
life-threatening condition which is caused by accumulation of metformin. Risk
factors include renal impairment, old age and doses over 2 g per day. The
estimated prevalence of lactic acidosis is one to five cases per 100,000
patients. Metformin is excreted unchanged in the urine, with the half-life
prolonged and renal clearance decreased in proportion to any decrease in
creatinine clearance. This may occur chronically in chronic renal impairment,
or acutely with dehydration, shock, and intravascular administration of
iodinated contrast agents, all of which have the potential to alter renal
function. Chloroquine Based on human studies, major adverse effects of
chloroquine which may limit dose are gastro-intestinal side effects (e.g.
nausea, diarrhoea) and hypoglycaemia. Furthermore, side-effects of chloroquine
include dizziness or blurred vision, sleep disturbances and headaches. Tumor
biopsy Possible adverse effect of a skin- and tumor biopsy include: - Bleeding
on the site of biopsy - Pain - Bruises - Infection Benefits Metformin and
chloroquine may have antitumor activity in IDH1/2-mutated cancers. Group
relatedness The anticancer activity of metformin may be bigger in IDH1-mutated
cancers than IDH2-mutated cancers. >90% of all IDH1/2-mutated intrahepatic
cholangiocarcinoma and chondrosarcoma have IDH1 mutations.