Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79

Contents lists available at ScienceDirect

Journal of Pharmacological and Toxicological Methods

journal homepage: www.elsevier.com/locate/jpharmtox

Validation and utility of the PhysioTel™ Digital M11 telemetry implant

for cardiovascular data evaluation in cynomolgus monkeys and
Beagle dogs
Jason S. Cordes a,⁎, Jonathan R. Heyen b, Marlo L. Volberg a, Nancy Poy c, Steven Kreuser a,
Ahmed M. Shoieb a, Jill Steidl-Nichols a
a
Pfizer, Eastern Point Road, Groton, CT, USA
b
Pfizer, 10777 Science Center Drive, San Diego, CA, USA
c
Research Surgery and Veterinary Consulting, San Francisco Bay Area, CA, USA

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: The cardiovascular liability of candidate compounds can be evaluated by a number of methods in-
Received 24 August 2015 cluding implanted telemetry, jacketed telemetry and surface lead electrocardiogram (ECG). The utility of the new
Received in revised form 15 December 2015 PhysioTel™ Digital M11 cardiovascular telemetry implant was evaluated in monkeys and dogs.
Accepted 27 January 2016 Methods: Eight monkeys and dogs (4 males and 4 females per species) were implanted with the M11 device uti-
Available online 1 February 2016
lizing a femoral blood pressure catheter and periosteal ECG leads. The signal quality of the ECGs was determined
as a percentage of software-matched waveforms and as a percentage of signal loss during the recording periods.
Keywords:
Telemetry
To investigate sensitivity for detecting changes in QT/QTc and HR/BP, moxifloxacin and doxazosin were admin-
PhysioTel™ Digital istered to monkeys and dogs implanted with the M11 device. Additionally, histopathological evaluation of the
M11 implant site was completed.
JET Results: For both monkey and dog, the percentage of recognizable waveforms was high (65% and 85%, respective-
Cardiovascular ly), while the average amount of signal loss was low (1% and 3%, respectively), indicating that the M11 implants
Moxifloxacin delivered data of sufficient quality. In monkeys, moxifloxacin (90 mg/kg) induced QT and QTc prolongation up to
Doxazosin 22 and 12 ms, respectively, while at 30 mg/kg in dogs, the maximal increases in QT and QTc were 13 and 16 ms,
Safety pharmacology
respectively. Doxazosin (1.5 and 1.0 mg/kg) produced HR increases up to 35 and 29 bpm with decreases in blood
QTc prolongation
pressure up to −14 and −26 mm Hg in monkeys and dogs, respectively. The histopathological impact of the im-
plant, catheter and biopotential leads was limited to expected minor local inflammatory changes as assessed at
necropsy and with microscopic examination.
Discussion: Based upon the results of this study, the PhysioTel™ Digital M11 is a suitable technology for assessing
cardiovascular parameters in monkeys and dogs, and because of the size and limited invasiveness of the implant,
is well positioned for use on toxicology studies.
© 2016 Elsevier Inc. All rights reserved.

1. Introduction Harmonization of Technical Requirements for Registration of Pharma-

Evaluation of cardiovascular (CV) endpoints on toxicology studies is
typically performed using surface lead electrocardiograms (ECG), and
on occasion, with a tail cuff measurement for blood pressure (BP). How-
ever, with the advent of jacketed external telemetry (JET) with a BP
add-on, more robust cardiovascular evaluation can be implemented
on toxicology studies to better characterize the CV liability of candidate
compounds. Collecting CV data as an add-on, and potentially replacing a
standalone study, is in alignment with The International Conference on
⁎ Corresponding author.
E-mail address: Jason.s.cordes@pfizer.com (J.S. Cordes).
ceuticals for Human Use (ICH) guidelines (ICH, 2000, 2009) and further
promotes the principles of the 3Rs (Replacement, Reduction, Refine-
ment) for animal use. JET has been successfully implemented and is
widely used by the pharmaceutical industry, and is especially useful
for testing CV risk for biologics for which the use of telemetry colony an-
imals is not an efficient option. However, JET does have challenges pri-
marily associated with jacketing and ECG surface lead placement that
make alternative technologies attractive.
The purpose of this investigation was to evaluate the newly released
M11 implant with PhysioTel™ technology [Data Sciences International
(DSI)] as a potential alternative to JET. Several features of the implant
make it suitable for use on toxicology studies, while offering advantages
over JET. Like the PA-C40 or PA-C10 implants (DSI) used with JET, the

http://dx.doi.org/10.1016/j.vascn.2016.01.006
1056-8719/© 2016 Elsevier Inc. All rights reserved.
 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79
M11 is small and only requires a minor surgery for implantation. How-
ever, the M11 does not require the use of jackets for collection of CV data
which is an added animal welfare benefit. Wearing the jackets can
induce animal stress and has been shown to elevate hemodynamic pa-
rameters in monkeys (Cordes, Harris, Holloway, Foote, & Steidl-Nichols,
2011) and dogs (data unpublished). Furthermore, the M11 implants
provide surgically implantable bipolar leads for collection of ECGs in ad-
dition to the single pressure catheter for blood pressure monitoring.
These advancements present a clear advantage in terms of consistency
and reliability, over JET, which requires surface lead ECG collection.
Therefore, the M11 technology offers potential improvements to animal
welfare and data quality, as well as significant time and resource savings
that can be realized through the elimination of jackets while maintain-
ing the use of minor surgery to implant.
Prior to adoption of any new implanted telemetry technology for
pharmaceutical testing, it is important to assess the reliability of the
hardware, data quality and assay sensitivity, as well as the impact of
the implant on animal health and pathology (especially for toxicology
studies). In these studies with the M11 implant, data quality, detection
of hemodynamic and ECG changes with administration of positive con-
trol agents (moxifloxacin and doxazosin), and impact to the histopath-
ological examination were evaluated.
2. Methods
2.1. Study design
This study encompassed method validation for the M11 implant and
positive control testing in a cohort of monkeys and dogs at two different
test sites. Animals surgically implanted with the transmitter were given
vehicle and positive control compound in crossover designed studies
(n = 8 animals/group). Based upon the pharmacokinetic profiles of
each compound, at least 5 days (N7 half lives) were permitted between
dose administrations to ensure clearance of the compounds from sys-
temic circulation. The alpha-1 adrenergic receptor antagonist doxazosin
(Pfizer, Inc.) was administered as a single oral dose to monkeys
(1.5 mg/kg in 20% PEG and water) and dogs (1 mg/kg in 0.5% methylcel-
lulose and water). The fluoroquinolone antibiotic moxifloxacin
(Sequoia Research Products, Ltd.) was administered as a single oral
dose to monkeys (90 mg/kg) and dogs (30 mg/kg) in a vehicle of 0.5%
methylcellulose, 0.1% polysorbate 80 and water.
2.2. Animals
All experiments involving animals were conducted as per the guide-
lines and study protocols reviewed and approved by Pfizer Institutional
Animal Care and Use Committee.
Four male and four female cynomolgus monkeys (Macaca fascicularis)
of Mauritius origin (SNBL and Covance) were used for the study, were at
least 2.5 years old and ranged in body weight between 4.3 and 9.0 kg.
Monkeys were housed in stainless steel cages and pair-housed at all
times with the exception of telemetry days. A standard diet of pelleted
food (Certified Primate Diet 5K91) supplemented with vegetables and/
or fruit was provided twice daily at approximately 7:00 AM (prior to
dosing) and 1:00 PM (4 hours postdose (HPD)) and remained with
the animals overnight. Water was provided ad libitum. Environmental
conditions were maintained between 16 and 25 °C, 30–70% humidity,
a minimum of 12 air changes/h and a 12-hour light/dark cycle (6 AM/
6 PM lights on/off).
Four male and four female Beagle dogs (Canis lupus familiaris) of
Marshall Farms origin were used for this study, were at least 1 year
old and ranged in body weight between 8.2 and 10.2 kg. Dogs were
housed in 4 ft. × 6 ft. runs and pair-housed at all times with the excep-
tion of telemetry days. A standard diet of dry food (Certified Purina Diet
5L66) was provided once daily at approximately 2:00 PM (5 HPD) and
remained with the animals overnight. Water was provided ad libitum.
73
Environmental conditions were maintained between 16 and 25 °C,
30–55% humidity, a minimum of 12 air changes/h and a 12-hour light/
dark cycle (6:30 AM/6:30 PM lights on/off).
Eight (n = 8) animals of each species were chosen for these studies
to provide a relevant comparison to internal historical data, as eight
animals are commonly monitored as part of a toxicology CV add-on or
a traditional standalone Safety Pharmacology telemetry study in our
facility.
2.3. M11 implant
The M11 is a PhysioTel™ Digital implant (Fig. 1A) intended for use
with large animals and is smaller (11 cm3) and lighter (13.7 g) than
the L11 or D70-PCTP transmitters from DSI, but slightly larger than
the PA-C10 or PA-C40 BP implants used with JET. The M11 implants
used on this study had 1 pressure catheter (which was 1.4 mm in diam-
eter and 35 cm in length for monkeys, and 0.7 mm in diameter and
15 cm in length for dogs), biopotential leads for ECG collection, and tem-
perature and activity sensors. A comparison of the PhysioTel™ Digital
M11 implant, D70-PCTP and JET BP PA-C40 transmitter specifications
can be found in Supplementary Table 2.
2.4. Surgical procedures
2.4.1. Monkey
Monkeys were anesthetized with isoflurane in oxygen following
premedication with buprenorphine, meloxicam, glycopyrrolate and
ketamine (sedation). Intra-operative administrations included intrave-
nous (i.v.) fluids, cefazolin (i.v.), and an external heat source. Routine
anesthetic monitoring was performed, which also included direct
blood pressure measurement (via telemetry device). An intermuscular
pocket was created between the muscles of the internal oblique and
transversus abdominus muscles through aponeurosis incision. The
transmitter was inserted and secured with nylon sutures with the
antenna perpendicular to midline in the lower abdominal muscle fascia
of the external oblique. The systemic blood pressure catheter was
tunneled and inserted 10 cm into the left femoral artery with the tip
into the external iliac artery. ECG biopotential leads were secured in a
lead 2 configuration to the costal periosteum with nonabsorbable braid-
ed polyester sutures. Bupivicaine was used as a splash block in the sur-
gical sites prior to three layer closure with absorbable polydioxanone in
the fascia, and poliglecaprone 25 to appose muscle and dermis. Animals
were placed on scheduled analgesics (buprenorphine and meloxicam)
in the postoperative period and completed an uneventful recovery. A
full physical examination and complete blood count (CBC)/blood chem-
istry analysis was performed prior to release for study use.
2.4.2. Dog
Dogs were anesthetized with isoflurane in oxygen following
premedication with acepromazine, meloxicam, glycopyrrolate and
buprenorphine. Intra-operative administrations included intravenous
(i.v.) fluids, cefazolin (i.v.), and external heat source. Routine anesthetic
monitoring was performed which also included direct blood pressure
(via telemetry device). An intermuscular pocket was created for trans-
mitter placement between the muscles of the external and internal
abdominal oblique along the left lateral flank. The antenna was directed
across cranially in the fascia layers following the body wall using
blunt dissection for insertion. The systemic blood pressure catheter
was tunneled and inserted 15 cm into the left femoral artery with
the catheter tip residing in the illiac artery. ECG biopotential leads
were secured in lead 2 configuration to the costal periosteum with non-
absorbable braided polyester sutures. Bupivicaine was used as a
splash block in the surgical sites prior to three layer closure with absorb-
able Polydioxanone in the fascia, and poliglecaprone 25 to appose
muscle and dermis. Animals were placed on scheduled analgesics
(buprenorphine and meloxicam) in the postoperative period and
74 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79
Fig. 1. A) Photograph of the PhysioTel™ Digital M11 implant (courtesy of DSI) showing the antenna, blood pressure catheter and 2 bipolar leads. B) Five seconds of representative BP (top)
and ECG (bottom) signals acquired from a monkey instrumented with the M11 implant.
completed an uneventful recovery. A full physical examination and CBC/
blood chemistry analysis prior was performed prior to release for study
use.
2.5. Data collection and analysis
Animals were acclimated to the telemetry cages or runs at least
overnight prior to data collection. Telemetered data were acquired
continuously beginning approximately 1 h prior to dosing through ap-
proximately 22 HPD (monkeys) or 24 HPD (dogs) while the animals
were freely moving in cages/runs. Monkey telemetry recording utilized
2 transceivers (TRX-1, DSI) per group of 4 cages, while dog telemetry re-
cording used 1 TRX-1 per animal. Room interruptions were limited to
twice daily at the times of dosing and ~ 4 or 5 HPD for feeding. Blood
pressure, ECG, temperature and activity data were recorded at a sam-
pling rate of 500, 500, 10 and 10 Hz, respectively, using Ponemah Phys-
iology Platform P3 Plus, version 5.2 (Data Sciences International). Post-
acquisition analysis of the data were performed with Ponemah ECGpro
version 5.0 (Data Sciences International) or ecgAUTO version 2.4.0.20
(emka TECHNOLOGIES).
Data quality was assessed by examining the data for signal loss and
by utilizing a measurement of success. The 1 minute mean data lines
with zero values (indicative of signal dropout, maximum negative or
positive values (rail) or unreadable ECG waveforms) were considered
a marker of signal loss. Signal loss for each recording session was deter-
mined as a percentage of 1 minute data loss versus the total number of
1 minute data collected. To assess the ECG data success, the percentage
of the recognizable ECG waveforms compared to the total number of R-
waves in each 1 minute segment was determined.
Analyzed data were summarized as 1 minute mean values and
transferred to a Microsoft Excel analysis workspace where data lines
were automatically processed to provide contiguous 15 minute mean
values. The 15 minute mean data were further averaged by binning
into large time periods (super-intervals) for statistical analysis [Interval
1: ~1 to 4 HPD; Interval 2: ~6 to 9 HPD; Interval 3: ~10 to 21 HPD]. This
method of data analysis has been described previously (Sivarajah et al.,
2010).
An individual correction factor was generated from the vehicle con-
trol data for each animal as a method of normalizing QT-interval over a
range of RR-intervals (monkey) or heart rates (dog). For each animal,
correlation between 1 minute mean RR-interval or HR values and asso-
ciated QT-interval values was determined using linear regression. The
slope of this linear regression was used as the individual correction
factor for the associated animal for all treatments. This individual cor-
rection factor was used to calculate the corrected QT-interval (QTc)
using the following equations:
Monkey : QTc ¼ QT interval corrected for heart rate
¼ QT−½ðRR  550Þ  ðslopeÞ
Dog : QTc ¼ QT interval corrected for heart rate
¼ QT−½ðHR  75Þ  ðslopeÞ:
Statistical analysis was performed on combined sexes using SAS for
each of these collected or derived parameters (HR and RR-, PR-, QRS-,
QT-, and QTc-intervals; systolic, diastolic and mean blood pressure;
and activity). Each response was analyzed separately for each time peri-
od using repeated-measure analysis of variance (ANOVA) to investigate
differences due to treatment (vehicle and positive control compound)
while accounting for variation due to animal, super-interval and study
period (different days). Comparisons between positive control article
and vehicle were made for each animal. Statistical significance was
reported at the 5% level (p b 0.05). Additionally, a power analysis was
conducted for each species (n = 2 studies per species with the M11
technology) to determine the detectable differences with 80% power
for each of the analyzed parameters. A repeated-measure ANOVA was
used for this power analysis and the significance level was set at the
5% level (p b 0.05).
2.6. Histopathology
All eight monkeys underwent necropsy for assessment of the
pathological impact of the M11 device, catheter and ECG leads. Gross
observations were made at the time of necropsy. Sections of the femoral
artery at the point of catheter insertion and the surrounding muscle
bed/pocket, at the midline, and at the catheter tip were removed. The
samples were embedded in paraffin and sectioned at approximately
5 μm, and stained with hematoxylin and eosin (H&E), then submitted
for histopathology evaluation.
3. Results
3.1. Signal quality
In general, the ECG and BP signals were of high quality and were
recognizable by the ECG analysis software, as evidenced by the low
 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79
percentage of data loss and high percentage of matched waveforms. Ex-
ample ECG and BP signals from a monkey are shown in Fig. 1B and
demonstrate the typical quality of data acquired from the M11 implant.
The signal quality from the M11 implant was determined as a
percentage of the number of 1 minute data exclusions, versus the total
number of 1 minute data points collected from each animal over
the course of the studies. These data were averaged for each species
(n = 8 per study; 2 collections per animal) and plotted in Fig. 2A.
These data show that there are a minimal amount of 1 minute data seg-
ments with data loss, with monkeys and dogs averaging only 1% and 3%
data loss over the course of the study, respectively. This is similar to the
amount of data loss observed with Jacketed External Telemetry (JET, PA-
C10 used with monkey, PA-C40 used with dog) and traditional teleme-
try (D70-PCTP in both monkey and dog) hardware, which were b 3%.
Additionally, an assessment of the ECG quality was determined as
the percentage of the total number of recognizable ECG waveforms
versus the total number of waveforms collected (% success; Fig. 2B).
The M11 data analyzed from monkeys and dogs showed a high amount
of recognizable waveforms, with average percentages of 65% and 85%,
respectively, which is sufficient to describe the entire data collection
Fig. 2. The signal quality of the cardiovascular data collected from the PhysioTel™ Digital
M11 implant was expressed as a function of A) percent data loss and B) percent of
matched ECG waveforms (success) and compared to JET and telemetry (D70-PCTP)
technologies. The data are plotted as mean ± s.e.m. from vehicle control and drug
treated recording sessions from each species (n = 16 recording sessions for all except
dog D70-PCTP, for which n = 8 recording sessions). Data loss with the M11 technology,
as determined by the percentage of 1 minute data segments with no obtainable data,
was low for both monkeys and dogs (1% and 3%, respectively) and similar to that of JET
and D70-PCTP telemetry. The quality of the M11 ECG signal (success) was high for both
monkeys and dogs, with 65% and 85% of waveforms being matched, respectively. Data
collected using PhysioTel™ Digital, JET and D70-PCTP showed similar levels of success,
ranging from 65% to 94% across all three techniques. An asterisk (*) indicates a
statistically significant difference (p b 0.05, Student's t-test).
75
period. These values were similar to those collected with JET in each
species (72 and 82% for monkey and dog, respectively). When compar-
ing the M11 to the D70-PCTP technology, % success was similar for dog
(89%), but the %success rate was higher with the D70-PCTP technology
in monkeys (94%).
3.2. Positive control data
Moxifloxacin is a fluoroquinolone antibiotic that is known to
prolong the QT and QTc intervals (Chaves et al., 2006; Chen et al.,
2005; Sivarajah et al., 2010) and was used in this study to verify the
ability of the PhysioTel™ Digital technology to detect QT and QTc
interval prolongation in both monkeys and dogs (Figs. 3, 5A, Supple-
mentary Table 1). Following a single oral administration of 90 mg/kg
moxifloxacin to monkeys, there were no changes in heart rate (HR),
but QT interval prolongation (+16 to 22 ms) was observed throughout
the postdose monitoring period (~1–21 HPD). Upon correction for HR,
the QTc interval was also prolonged (+9 to 12 ms) over the same peri-
od (Figs. 3, 5A, Supplementary Table 1). Similarly, administration of
30 mg/kg moxifloxacin to dogs did not produce any change in HR,
but did prolong the QT interval (+7 to 13 ms) over the entire postdose
recording period. When corrected for HR, QTc interval prolongation
(+ 9 to 16 ms) was also evident. The statistical power analysis for
these studies indicated that the methods used here can detect small
changes in QTc in both monkeys and dogs (10 and 6 ms, respectively).
To determine the ability of the PhysioTel™ Digital technology to
detect changes in HR and BP, the alpha-1 adrenergic receptor antago-
nist, doxazosin, was used as a positive control (Figs. 4, 5B–C, Supple-
mentary Table 1). Oral administration of doxazosin at 1.5 mg/kg to
monkeys caused rapid decreases in systolic (SBP), diastolic (DBP) and
mean (MBP) blood pressure (−14, −8 and −11 mm Hg, respectively)
over the ~1–4 HPD period. Blood pressure returned to baseline levels by
~5 HPD. Heart rate increases (+35 bpm) were most pronounced during
the ~1–4 HPD period, but were sustained at smaller magnitudes (+11
to 17 bpm) from ~6–21 HPD. Doxazosin also produced similar results
when administered orally to dogs at 1 mg/kg (Figs. 4, 5B–C, Supplemen-
tary Table 1). In the dog, there were SBP, DBP and MBP decreases of
(−26, −7 and −14 mm Hg, respectively) during the ~1–4 HPD record-
ing period. The HR increases were maximal (+29 bpm) at 1–4 HPD, but
were also sustained at smaller magnitudes (+9 to 17 bpm) for the du-
ration of the postdose recording period. According to the statistical
power analysis conducted, small changes in HR and MBP were detect-
able at the 80% level in both monkeys and dogs (13 and 9 bpm; 7 and
5 mm Hg, respectively).
3.3. Histopathology
At the time of necropsy for the monkeys (3 months following im-
plantation), all implants were fully functional and there were no gross
abnormalities that could be associated with the periosteal ECG leads,
transmitter body or catheter.
At the point of catheter insertion, the femoral artery of 7/8 monkeys
showed minimal to mild trauma induced inflammatory (neutrophils,
macrophages, lymphocytes, and eosinophils) and degenerative changes
(neointimal hyperplasia/thickening of tunica intima), partial loss of in-
ternal elastic lamina, red blood cells, fibrin, and necrotic debris which
often enveloped the cross section of catheter within the lumen of the
blood vessel. The inflammation had extended to the tunica media in
2/8 animals (minimal). Some sections had soft tissue changes in the sur-
rounding muscle bed/pocket which were considered an extension of
trauma and were surrounding the artery. These consisted of granuloma-
tous inflammation composed of nodular mixture of epitheliod macro-
phages and variable numbers of multinucleated giant cells infiltrates
with cross sections of thick-wall greenish-transparent to refractile for-
eign materials (likely suture material), and occasional degeneration/
regeneration of skeletal muscle.
76 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79

Fig. 3. Timecourse graphs of the cardiovascular data collected from monkeys (A, C, E) and dogs (B, D, F) following oral administration of vehicle control or moxifloxacin. Data points are
expressed as 15 minute mean values (n = 8 for both monkeys and dogs) ± s.e.m. Dose administration occurred at 0 h, feeding occurred at 4–5 h and lights off and on at 9 and 21 h,
respectively. In both monkeys and dogs, moxifloxacin administration caused increases in QT and QTc when compared to vehicle control.

In the femoral artery (mid-catheter), minimal to mild intima
thickening/hyperplasia/proliferation was present in 8/8 monkeys, and
often affected one side of the vessel wall. The change was suggestive
of a “rubbing-like” irritation to the side of blood vessel making contact
with the catheter. There were minimal inflammatory infiltrates (neu-
trophils, macrophages, lymphocytes, and/or eosinophils), in 2/8 mon-
keys with small numbers of red blood cells, fibrin, fibroblasts, and
necrotic debris which often enveloped the cross section of catheter
within the lumen of the blood vessel. At the catheter tip, minimal to
mild intima proliferation was seen in 7/8 monkeys similar to observa-
tions seen in the midline section.
These minimal histopathologic findings above are expected and
consistent with well circumscribed, transmural arterial trauma and lo-
calized effects characteristic of a foreign body reaction (fibrosis, inflam-
mation) in the area in close contact with the body of the transmitter
device and catheter.
4. Discussion
According to the International Conference on Harmonization (ICH)
Guidelines S7A (ICH, 2000), the cardiovascular safety risk for potential
therapeutics must be assessed prior to First-In-Human (FIH) clinical
trials. However, this guidance does not mandate a stand-alone CV as-
sessment using telemetry for highly specific biologics. Instead, S7A
and additionally ICH S6 (ICH, 1997) and ICH S9 (ICH, 2009) recommend
the incorporation of CV endpoints onto toxicology studies with bio-
logics. Therefore, many across the industry have relied upon surface
lead ECG collection with restrained animals on toxicology studies to
satisfy the regulatory guidance. However, for some programs or indica-
tions, assessment of surface lead ECG may not provide sufficient CV risk
assessment. For these programs, it is advisable to incorporate more
thorough ECG and/or hemodynamic assessments. It is for these reasons
that many companies have adopted the use of JET add-ons (with or
without blood pressure implants) for toxicology studies to allow more
robust CV data collection and risk assessment.
JET add-ons have been successfully utilized as a supplement to or re-
placement for stand-alone Safety Pharmacology studies to support
programs spanning many therapeutic areas (Vargas, Amouzadeh, &
Engwall, 2013). The quality, sensitivity and reproducibility of JET ECG
and blood pressure data have been demonstrated and JET is widely ac-
cepted across the industry (Chui et al., 2009; McMahon, Mitchell, Klein,
Jenkins, & Sarazan, 2010). However, there are some aspects of JET that
present areas for potential improvement.
Based on our experience, wearing jackets does cause some level of
stress to the animals, as both monkeys (Cordes et al., 2011) and dogs
(data not published) have shown small increases in basal hemodynamic
parameters. Additionally, outfitting animals with jackets can prove to be
a difficult and time consuming activity, particularly with conscious
monkeys, where using 2–3 technicians to safely jacket each animal is
advisable. Moreover, jacket and hardware destruction is not uncommon
and creates an added cost for replacement equipment as well as a risk
for significant amounts of data loss subsequent to hardware destruction
or electrode removal. Skin irritation and consistent placement and ad-
herence of surface lead electrodes are also issues that can negatively im-
pact the data quality. Because of these issues, we investigated the use of
the PhysioTel™ Digital M11 implant as an alternative to JET.
The M11 implant offers the same qualities as JET with blood pressure
monitoring (high data quality, ability to record for extended durations,
minor surgery), and removes the issues associated with the jackets.
The M11 implant was successfully implanted in 8 monkeys and dogs.
Although still a minor surgery, both the implant procedure and
the recovery period are longer in duration than for the C10 or C40
 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79 77

Fig. 4. Timecourse graphs of the cardiovascular data collected from monkeys (A, C, E) and dogs (B, D, F) following oral administration of vehicle control or doxazosin. Data points are
expressed as 15 minute mean values (n = 8) ± s.e.m. Dose administration occurred at 0 h, feeding occurred at 4–5 h and lights off and on at 9 and 21 h, respectively. In both
monkeys and dogs, doxazosin administration caused increases in HR and decreases in SBP and DBP when compared to vehicle control.

transmitters used with JET. The surgery for JET BP implantation takes
about 40 min, while the M11 implant with ECG leads can take up to
80 min. The tunneling of ECGs also requires a few additional days of
postoperative monitoring. However, if the primary risk being investi-
gated is related to blood pressure, ECG leads are not necessary which
can bring the implant time closer to that of JET BP. Although surgical
investment is higher, by eliminating the need for jackets and external
hardware, animal welfare is improved and considerable time and cost
savings can be realized during the in-life stage through the elimination
of the jacket acclimation phase (2 weeks at our facility), many man
hours jacketing and unjacketing the animals, and time spent organizing
jackets and hardware during the study. Additionally, instead of adher-
ent surface lead ECG placement, the M11 implant utilizes periosteal
lead placement which improves the week to week consistency and
quality of the ECG data, and eliminates the possibility of significant
data loss due to ECG removal or poor adherence. The ability to detect
small changes in ECG parameters (e.g. 5–10 ms change in QTc) is depen-
dent upon the quality of the signal and reproducibility over the course of
the study. These data quality improvements also reduce the amount of
time needed to analyze and overread the data. With these improve-
ments in focus, the M11 implant is well-positioned to serve as a suitable
add-on technology in the same capacity as JET. The purpose of this study
was to evaluate the PhysioTel™ Digital M11 implant in terms of data
quality, ability to detect hemodynamic and ECG changes as well as as-
sessment of any impact to local pathology.
This study showed using the M11 implant consistently produced re-
liable, high quality, ECG and blood pressure signals from both monkeys
and dogs. The appearance of the ECG and blood pressure signals were
generally acceptable and of sufficient quality to analyze appropriately,
with minimal amounts of noise in the tracings except during times of
extreme animal movement. There were very few 1 minute segments
during the 22–24 hour recording periods where there was complete
data loss, demonstrating that the signal transmission from the implant
to the transceiver and acquisition system is reliable, and in any 1 min
there were likely to be recognizable ECG waveforms. Over the course
of the recording period, on average, data loss only accounted for 1%
and 3% (monkeys and dogs, respectively) of the total 1 minute data
points collected (an average of 1450 1-minute data points per collec-
tion). This finding is consistent with JET and D70-PCTP experience,
where data loss due to transmission interruption is minimal. Addition-
ally, the data loss segments tended to occur sporadically through the
data sets or during times of high activity (such as during dosing or
bleeding activities) and did not affect data interpretation.
To understand the quality of the M11 ECG signal, the percentage of
recognizable or software-matched ECG traces was used as a marker.
The amount of recognizable M11 ECG waveforms was 65 and 85% for
monkeys and dogs, respectively. This is consistent with previous studies
using JET (McMahon et al., 2010) and is similar to our own experiences
with JET surface lead ECG in both monkeys and dogs. We did find that
the percent of successful ECG waveform matches was lower for data col-
lected from the M11 transmitters compared to the D70-PCTP transmit-
ters for monkey, which is not unexpected as monkeys behavior and
movement in the cage would have a bigger impact on the quality of
data generated from periosteal leads (M11) versus leads attached
directly to the heart (D70-PCTP). Greater than 90% ECG waveform rec-
ognition is commonplace with both dogs and monkeys using the D70-
PCTP. Nonetheless, the M11 ECG waveforms obtained from the perios-
teal leads were of consistent morphology from collection to collection,
which is not the case with JET surface electrode placement where mor-
phology can shift from one collection to the next due to inconsistent
78 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79
Fig. 5. Changes in cardiovascular parameters following oral administration of moxifloxacin
(A) and doxazosin (B, C) to monkeys and dogs. Data were compared to vehicle control for
super-intervals 1, 2 and 3 (~ 1–4, 6–9, and 10–21 HPD time periods, respectively).
Statistically significant changes (p b 0.05) are indicated with *. The changes detected
with PhysioTel™ Digital technology were comparable to changes observed with
historical data collected with traditional internal telemetry devices from DSI (D70-PCT
or -PCTP) or Konigsberg Instruments (T27F). Doxazosin dog telemetry data was adapted
from previously published data (Sivarajah et al., 2010). n = 8 for PhysioTel™ Digital
studies, n = 8 for NHP telemetry, n = 4 for dog telemetry.
placement of the leads. The blood pressure catheter insertion technique
and final placement with the M11 was the same technique as used with
JET, and thus it was not surprising that blood pressure waveforms were
of similar quality to those generated with JET.
Because of the heightened awareness and sensitivity surrounding
QTc prolongation and the potential for fatal arrhythmia such as Torsades
de Pointes, it is of utmost importance that the preclinical models used to
assess effects on ECGs are shown to be sensitive for detecting effects on
QT and QTc. Moxifloxacin is a commonly used positive control for
demonstration of QT and QTc both preclinically and clinically (Chen
et al., 2005; Darpo, Nebout, & Sager, 2006; Shah, 2005). In this study,
the M11 device using periosteal ECG leads was capable of detecting QT
and QTc prolongation in both monkeys and dogs. Following a single
oral administration of 90 mg/kg to monkeys and 30 mg/kg to dogs,
moxifloxacin produced QT and QTc increases that were of similar magni-
tude to internal and published data (Chaves et al., 2006; Chen et al.,
2005; Sivarajah et al., 2010) using different telemetry hardware. These
data also showed that the statistical power in each of these studies was
sufficient for detecting small changes in QTc (approximately 6–10 ms).
Although QTc prolongation may garner the most attention, detecting
changes in HR and blood pressure preclinically are equally important
to ensure patient safety. In this study, the alpha-1 adrenergic receptor
antagonist, doxazosin, was used as a positive control for detecting he-
modynamic changes. Oral administration of doxazosin at 1.5 mg/kg
to monkeys and 1 mg/kg to dogs caused rapid decreases in blood pres-
sure and increases in HR that were consistent with previous data gener-
ated in monkeys and dogs using standard telemetry devices. Again,
these data indicate that the statistical power of these studies was
shown to be sufficient for detecting small changes in heart rate (approx-
imately 9–13 bpm) and blood pressure (approximately 5–7 mm Hg)
parameters.
The M11 device is intended for use to support a more robust cardio-
vascular evaluation in a toxicology study. Therefore, it was important to
assess the potential impact of the hardware on pathological endpoints
to ensure that the hardware would not confound interpretation of tox-
icology studies. It has been demonstrated that there are only minor, lo-
calized effects to pathology with JET blood pressure (PA-C10 LA) or fully
implantable (D70-PCTP) telemetry implants (Pfizer internal data, Baird
et al., 2013). In this study, 8 monkeys were examined at necropsy for
gross abnormalities related to the implant, catheter and subcutaneous
ECG leads, with no findings noted. Additionally, microscopic examina-
tion of the implant site, and cross-sections of the catheter throughout
the femoral artery were performed. Importantly, all effects were mini-
mal to mild and were localized to the areas near the transmitter device
and catheter. Only minimal to mild changes were noted, which included
inflammatory (neutrophils, macrophages, lymphocytes, and eosino-
phils) and degenerative changes (neointimal hyperplasia/thickening
of tunica intima), partial loss of internal elastic lamina, red blood cells,
fibrin, and necrotic debris. These changes were not unexpected based
on results from previous investigations (Baird et al., 2013) and are con-
sistent with expected localized effects (fibrosis, inflammation) in areas
in close contact to the transmitter device. These results provide confi-
dence that the M11 device is suitable for use on toxicology studies as
these minimal, localized findings at the implant site would not interfere
with the pathology interpretation.
5. Conclusion
In summary, the newly available PhysioTel™ Digital M11 telemetry
implant utilizing periosteal ECG leads and a blood pressure catheter
is an acceptable technology for assessing CV effects of new pharmaceu-
tical candidates on toxicology studies. By removing the need for jackets,
the M11 device offers advantages over JET in terms of animal welfare as
well as resource savings. The telemetry data from the M11 device was of
sufficient quality and reliability, with minimal instances of data loss and
high percentages of recognizable waveforms. Expected moxifloxacin-
induced QT and QTc prolongation, as well as doxazosin-induced chang-
es in HR and blood pressure, were detectable in both monkeys and dogs
implanted with the M11 device. Additionally, the minor surgery, mini-
mal impact to local pathology and device longevity of at least 3 months,
make it suitable for use in the toxicology setting.
 J.S. Cordes et al. / Journal of Pharmacological and Toxicological Methods 79 (2016) 72–79
Acknowledgments
The authors wish to acknowledge Laura Ringer, Robert Barnes,
Stephen Foote, Dingzhou Li and Nicola Regan for technical contributions
and Data Sciences International for technological support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.vascn.2016.01.006.
References
Baird, T., Baile, M., Patrick, D. J., Moddrelle, D., Yoder, J., Gauvin, D. V., & Dalton, J. A. (2013).
Influence of surgically implantable telemetry solutions on in-life and post-mortem
toxicology endpoints. Journal of Pharmacological and Toxicological Methods, 67,
148–161.
Chaves, A. A., Keller, W. J., O'Sullivan, S., Williams, M. A., Fitzgerald, L. E., McPherson, H. E., ...
Briscoe, R. J. (2006). Cardiovascular monkey telemetry: Sensitivity to detect QT inter-
val prolongation. Journal of Pharmacological and Toxicological Methods, 54, 150–158.
Chen, X., Cass, J. D., Bradley, J. A., Dahm, C. D., Sun, Z., Kadyszewski, E., ... Zhou, J. (2005). QT
prolongation and proarrhythmia by moxifloxacin: Concordance of preclinical models
in relation to clinical outcome. British Journal of Pharmacology, 146, 792–799.
Chui, R. W., Fosdick, A., Conner, R., Jiang, J., Bruenner, B., & Vargas, H. (2009). Assessment
of two external telemetry systems (PhysioJacket™ and JET™) in beagle dogs with te-
lemetry implants. Journal of Pharmacological and Toxicological Methods, 60, 58–68.
Cordes, J. S., Harris, P. D., Holloway, J. B., Foote, S. C., & Steidl-Nichols, J. (2011, July–
August). Implementation of JET™ (Jacketed External Telemetry) for cardiovascular
79
monitoring in cynomolgus monkeys. Journal of Pharmacological and Toxicological
Methods, 64(1), e47.
Darpo, B., Nebout, T., & Sager, P. T. (2006). Clinical evaluation of QT/QTc prolongation and
proarrhythmic potential for nonantiarrhythmic drugs: The International Conference
on Harmonization of Technical Requirements for Registration of Pharmaceuticals
for Human Use E14 Guideline. Journal of Clinical Pharmacology, 46, 498–507.
Data Sciences International (2015/11/30). www.datasci.com
ICH (International Conference on Harmonization of Technical Requirements for
Registration of Pharmaceutical for Human Use) (1997, Julyl). Preclinical safety evalu-
ation of biotechnology-derived pharmaceuticals S6(R1). Available at http://www.
ich.org
ICH (International Conference on Harmonization of Technical Requirements for
Registration of Pharmaceutical for Human Use) (2000, Novemberl). Safety pharma-
cology studies for human pharmaceuticals S7A. Available at http://www.ich.org
ICH (International Conference on Harmonization of Technical Requirements for
Registration of Pharmaceutical for Human Use) (2009, Octoberl). Nonclinical evalua-
tion for anticancer pharmaceuticals S9. Available at http://www.ich.org
McMahon, C., Mitchell, A. Z., Klein, J. L., Jenkins, A. C., & Sarazan, R. D. (2010). Evaluation of
blood pressure measurement using a miniature blood pressure transmitter with
jacketed external telemetry in cynomolgus monkeys. Journal of Pharmacological and
Toxicological Methods, 62, 127–135.
Shah, R. R. (2005). Drugs, QT interval prolongation and ICH E14: The need to get it right.
Drug Safety, 28(2), 115–125.
Sivarajah, A., Collins, S., Sutton, M. R., Regan, N., West, H., Holbrook, M., & Edmunds, N.
(2010). Cardiovascular safety assessments in the conscious telemetered dog:
Utilisation of super-intervals to enhance statistical power. Journal of Pharmacological
and Toxicological Methods, 62, 12–19.
Vargas, H. M., Amouzadeh, H. R., & Engwall, M. J. (2013). Nonclinical strategy consider-
ation for safety pharmacology: Evaluation of biopharmaceuticals. Expert Opinion on
Drug Safety, 12(1), 91–102.