24-Hour ambulatory blood pressure monitoring in patients with malignancies enrolled in phase I oncological studies

In the past few years, novel anti-cancer agents have been introduced that are
able to inhibit tumor growth by targeting the formation of new blood vessels
(angiogenesis) of the tumor. It is hypothesized that if a tumor is no longer
able to generate new blood support, it loses its ability to grow, and might
even succumb altogether.[1] Drugs aimed at inhibiting the angiogenetic process
have very specific pharmacological mechanisms of action (*targeted therapy*),
and do not show the side-effects of the classical cytotoxic agents. However,
elevated blood pressure (hypertension) is often observed and is treatment
limiting. Although anti-hypertensive drugs often are able to treat the observed
hypertension, in some cases the condition can become life-threatening
(malignant hypertension) and lead to irreversible damage to eyes, kidneys,
lungs and/or brains. Moreover, poorly controlled hypertension can lead to
serious cardiovascular events, such as cardiac ischaemia and infarction.[2]
These side-effects can cause some patients to stop drug treatment prematurely
or continue on a lower dose, possibly resulting in decreased treatment
efficacy. Several mechanisms are hypothesized to be involved in the occurence
of this side effect, often involving vascular endothelial growth factor
(VEGF).[3] In years to come, new inhibitors of the VEGF-transduction route will
become available to the clinic that may also show hypertension as major side
effect.[4] A good insight into, and -where possible- control of this side
effect is therefore of significant importance. We aim to develop a
pharmacokinetic-pharmacodynamic model for the relationship between
antiangiogenic drug exposure and development of hypertension. When such a model
is available, the model can be used to optimize dosing schedules to minimize
the probability of developing hypertension, and to identify patients
particularly at risk for this side-effect. It is known however, that the blood
pressure varies during the day in healthy subjects, in a periodic fashion
(circadian rhythm). This obfuscates the PK-PD relationship, and a base signal
describing these circadian fluctuations should therefore be included into the
PK-PD model. As blood pressure and circadian variation in healthy volunteers
most likely are not comparable to those of cancer patients, these data have to
be recorded from the intented population, i.e. cancer patients.

Population pharmacokinetic-pharmacodynamic models incorporating the circadian
rhythm built on 24-hour ambulatory blood pressure measurements (ABPM) have been
described before in several populations, e.g. in patients with
hypertension[5,6], but not yet in patients with cancer. These models generally
consist of multiple sinusoidal functions. During the night, in most patients
blood pressure decreases >10% as compared to daytime (dippers), while in some
patients this decrease is not observed or even elevated blood pressures are
observed (non-dippers). Generally, blood-pressures are elevated during the
first few hours after arising, while throughout the day, external circumstances
can alter blood pressure as well. Through measurement of ABPM a mathematical
model of the circadian rhythm can be constructed, including both fixed effects
(parameters describing the population mean behaviour) and random effects (e.g.
interindividual variation in amplitude). The model can subsequently be used as
a base signal for the PK-PD model for hypertension associated with
anti-angiogenic therapy.

The aim of the study is to obtain a profile of the circadian rhythm in blood
pressure, in a representative population of cancer patients, thereby providing
a base signal for the pharmacokinetic-pharmacodynamic model for hypertension
associated with anti-angiogenic therapy.

Methodology
Monitoring will be performed using ABPM monitors, which are validated to
international standards (British Heart Association / the American Association
for the Advancement of Medical Instrumentation) and in routine use at the
Slotervaart Hospital. All ABPM monitors will be calibrated, and programmed to
record blood pressure readings every 30 minutes during the 24 hour cycle.
Monitors will be placed on the patient*s non-dominant arm. Patients are
instructed to keep the cuff and monitor on for 24 hours, and will be supplied
with activity diaries for recording approximate sleep time, exertional event
times and any other information pertinent to the study. It will be asserted by
the investigator or nursing staff that the cuff is placed on the arm with an
appropriate pressure, and that inflating the cuff does not present serious
inconveniences to the patient. Also, upon installment of the device, two
additional readings using a regular sphygmomanometer will be performed to
confirm the monitor is functioning.

Data recorded in the study will be converted to the appropriate data format by
the investigator. Mathematical models are fitted to the blood pressure data
using non-linear mixed-effets modeling software (NONMEM). Population parameters
are estimated, as well as corresponding inter-individual variability
descriptors. If possible, additional data provided by the patient such as sleep
times and exertional activities will be incorporated in the model as covariates
or in another appropriate statistical manner.

Burden:
Patients need to be equipped with a ABPM-device which monitors the blood
pressure at preprogrammed intervals (each half hour). This can lead to minor
inconveniences, such as localized increased pressure in the arm, or disturbance
of sleep.