Content uploaded by Shawn Michael Gage
Author content
All content in this area was uploaded by Shawn Michael Gage on May 24, 2020
Content may be subject to copyright.
Arterial reconstruction with human bioengineered
acellular blood vessels in patients with peripheral
arterial disease
Piotr Gutowski, MD, PhD,
a
Shawn M. Gage, BS,
b,c
Malgorzata Guziewicz, MD, PhD,
d
Marek Ilzecki, MD, PhD,
e
Arkadiusz Kazimierczak, MD, PhD,
a
Robert D. Kirkton, PhD,
b
Laura E. Niklason, MD, PhD,
b,f
Alison Pilgrim, MD,
b
Heather L. Prichard, PhD,
b
Stanislaw Przywara, MD, PhD,
e
Rabih Samad, MD, PhD,
a
Bill Tente, MS,
b
Jakub Turek, MD,
d
Wojcieh Witkiewicz, MD, PhD,
d
Norbert Zapotoczny, MD,
d
Tomaz Zubilewicz, MD, PhD,
e
and
Jeffrey H. Lawson, MD, PhD,
b,c
Szczecin, Wroclaw, and Lublin, Poland; Durham,
NC; and New Haven, Conn
ABSTRACT
Objective: Vascular conduit is essential for arterial reconstruction for a number of conditions, including trauma and
atherosclerotic occlusive disease. We have developed a tissue-engineered human acellular vessel (HAV) that can be
manufactured, stored on site at hospitals, and be immediately available for arterial vascular reconstruction. Although the
HAV is acellular when implanted, extensive preclinical and clinical testing has demonstrated that the HAV subsequently
repopulates with the recipient’s own vascular cells. We report a first-in-man clinical experience using the HAV for arterial
reconstruction in patients with symptomatic peripheral arterial disease.
Methods: HAVs were manufactured using human vascular smooth muscle cells grown on a biodegradable scaffold.
After the establishment of adequate cell growth and extracellular matrix deposition, the vessels were decellularized
to remove human cellular antigens. Manufactured vessels were implanted in 20 patients with symptomatic pe-
ripheral arterial disease as above-knee, femoral-to-popliteal arterial bypass conduits. After HAV implantation, all
patients were assessed for safety, HAV durability, freedom from conduit infection, and bypass patency for 2 years.
Results: Twenty HAVs were placed in the arterial, above-knee, femoral-to-popliteal position in patients with rest pain
(n ¼3) or symptomatic claudication (n ¼17). All HAVs functioned as intended and had no evidence of structural failure
or rejection by the recipient. No acute HAV infections were reported, but three surgical site infections were docu-
mented during the study period. Three non-HAV-related deaths were reported. One vessel developed a pseudoa-
neurysm after suspected iatrogenic injury during a balloon thrombectomy. No amputations of the HAV implanted
limboccurredoverthe2-yearperiod,andnoHAVinfections were reported in approximately 34 patient-years of
continuous patient follow-up.
Conclusions: Human tissue engineered blood vessels can be manufactured and readily available for peripheral arterial
bypass surgery. Early clinical experience with these vessels, in the arterial position, suggest that they are safe, have
acceptable patency, a low incidence of infection, and do not require the harvest of autologous vein or any cells from the
recipient. Histologic examination of tissue biopsies revealed vascular remodeling and repopulation by host cells. This first-
in-man arterial bypass study supports the continued development of human tissue engineered blood vessels for arterial
reconstruction, and potential future expansion to clinical indications including vascular trauma and repair of other size-
appropriate peripheral arteries. (J Vasc Surg 2020;-:1-12.)
Keywords: Peripheral arterial disease; Arterial reconstruction; Bioengineered blood vessel; HAV
From the Department of Vascular Surgery and Angiology, Pomeranian Medical
University of Szczecin, Szczecin
a
; Humacyte, Inc,
b
and the Department of Sur-
gery, Duke University,
c
Durham; the Research and Development Centre,
Department of Vascular Surgery, General Hospital, Wroclaw
d
; the Clinic of
Vascular Surgery and Angiology, Medical University of Lublin, Lublin
e
; and
the Departments of Anesthesia and Biomedical Engineering, Yale University,
New Haven.
f
Author conflict of interest: J.H.L. is CEO of Humacyte, and is a paid employees of
Humacyte, including salary and stock options, and serves on the board of di-
rectors. R.D.K. and B.T. are paid employees of Humacyte, including salary and
stock options. H.L.P. is a paid employee of Humacyte and has stock options.
H.L.P. has patents licensed to Humacyte. L.E.N. is a founder and shareholder
of Humacyte. L.E.N.’s spouse has equity in Humacyte. L.E.N. serves on the
board of directors and has patents that are licensed to Humacyte and pro-
duce royalties for L.E.N. L.E.N. has received unrestricted gifts to support her
research at Yale. A.P. is a consultant paid by Humacyte. Payment includes
salary and stock options. S.G. has been a paid employee and consultant to
Humacyte, and owns shares in the company. N.Z., M.I., S.P., T.Z., W.W., M.G.,
and J.T. have been paid as investigators in the study. W.W. has been payed
a consulting fee. R.S. has no conflicts of interest to disclose.
Presented at the 2017 Vascular Annual Meeting of the Society for Vascular Sur-
gery, San Diego, Calif, May 31-June 3, 2017.
Additional material for this article may be found online at www.jvascsurg.org.
Correspondence: Jeffrey H. Lawson, MD, PhD, Adjunct Professor of Surgery and
Pathology Department of Surgery, Duke University, Chief Executive Officer,
Humacyte Incorporated Durham, NC 27713 (e-mail: lawson@humacyte.com).
The editors and reviewers of this article have no relevant financial relationships to
disclose per the JVS policy that requires reviewers to decline review of any
manuscript for which they may have a conflict of interest.
0741-5214
Copyright Ó2019 by the Society for Vascular Surgery. Published by Elsevier Inc.
https://doi.org/10.1016/j.jvs.2019.11.056
1
Arterial bypass for peripheral arterial disease (PAD) and
vascular trauma has progressed over the past 70 years.
1
Early reports highlighted the use of both autologous vessels
and synthetic conduits as options for vascular repair.
2,3
Although vascular reconstruction has evolved, the need
for suitable bypass conduit has remained an important
consideration for every surgical case. Many conduit tech-
nologies have been developed for vascular bypass,
including the harvesting of autologous vessels, as well as
polyesters, xenografts, fresh/frozen homografts, and
expanded polytetrafluoroethylene. Although each of these
technologies has been important in the advancement of
modern vascular surgery, each material has specificlimita-
tions when used in surgical vascular reconstruction.
Autologous saphenous vein, which is considered the
gold standard for vascular repair, requires prolonged
operating time and produces harvest site pain, as well
as an increased risk of wound infection from the donor
site. Furthermore, saphenous vein quality can vary by pa-
tient and can be constrained by vein scaring, vascular dis-
ease, and varicosities.
4
Conduits made of synthetic
materials suffer from a lack of true biocompatibility
and a prolonged risk of infection. Xenografts have
elevated rates of early thrombosis and medial calcifica-
tion,
5
and cryopreserved veins have a persistent risk of
immune recognition.
6
Human umbilical veins, devel-
oped as bypass materials several decades ago, do not
show appreciably improved outcomes as compared
with synthetic conduits in recent clinical studies.
7
In an attempt to address the need for immediately
available, nonantigenic human vascular material to be
used for vascular reconstruction, we have devised a
method to grow human vessels in vitro using human
vascular cells, that are cultured on a biodegradable scaf-
fold. These vessels are then rendered acellular by a decel-
lularization process that gently rinses antigenic cellular
material from the vessel, preserving the extracellular ma-
trix proteins and mechanical integrity of the conduit,
resulting in a human acellular vessel (HAV).
8,9
We have reported the use of these vessels in a phase I/II
clinical study using the HAV as arteriovenous access in
patients with end-stage renal disease.
10
To evaluate the
early clinical usefulness of the HAV as an arterial conduit,
we conducted this study in patients with symptomatic
occlusion of the superficial femoral artery (SFA). To our
knowledge, this phase II study is the first assessment of
a completely human bioengineered blood vessel as an
arterial bypass conduit in the peripheral circulation.
METHODS
Production of HAVs. HAVs were 6 mm in diameter and
35 to 42 cm in length. Human vascular smooth muscle
cells were derived from deceased organ and tissue do-
nors, meeting eligibility requirements for all relevant
communicable diseases. After smooth muscle cell isola-
tion and expansion, cells were seeded onto degradable
polymer scaffolds that were contained within flexible,
single-use bioreactors. Developing vessels were sub-
jected to pulsatile cyclic distension and then were
decellularized to remove immunogenic cellular antigens
while preserving the comparatively nonimmunogenic
extracellular matrix constituents. The process to grow the
HAVs takes approximately 10 weeks (Fig 1).
11-13
Study design. We performed a prospective, open-label,
single treatment arm, multicenter pilot study. The pri-
mary objectives of the study were to evaluate the safety
of the HAV as an above-knee femoral-to-popliteal bypass
graft, and to determine the patency (primary, primary
assisted, and secondary) over 24 months. Patients with
PAD requiring above-knee peripheral bypass surgery
were screened within 14 days of the planned operation.
Patients were between the ages of 54 and 79 years (in-
clusion criteria age 18-80 years) with a projected life ex-
pectancy of at least 2 years, and with claudication at a
distance of more than 200 m or with ongoing rest pain.
All patients had a total occlusion segment of the SFA
(Inter-Society Consensus for the Management of Pe-
ripheral Arterial Disease II type B [n ¼5] and type C le-
sions [n ¼15]
14
) with adequate proximal inflow (distal
external iliac artery, common femoral artery, or proximal
SFA) and distal outflow (SFA or above-knee popliteal ar-
tery), with at least two-vessel runoff below the knee to
the ankle. Appropriate anatomy was assessed and
confirmed preoperatively with conventional or
computed tomography angiography. The HAV was used
as a conduit for bypass of an arterial occlusive lesion
within the SFA, which was not amenable to endovascular
therapy, and in which suitable autologous conduit was
not available for bypass.
A total of 20 eligible patients were enrolled at 3 clinical
sites and received the HAV implant (day 1). Patients were
followed for 2 years to assess the safety and efficacy of
the HAV in terms of patency, necessary graft interven-
tions, and relief of PAD symptoms of rest pain and clau-
dication. Six follow-up visits were performed in the first
year: immediately after surgery (days 5 and 15); at weeks
6, 12, and 26; and at month 12. Two follow-up visits were
performed in the second year (months 18 and 24).
ARTICLE HIGHLIGHTS
dType of Research: Prospective, open-label, single-
arm treatment, multicenter pilot study
dKey Findings: All human acellular vessels functioned
as intended and had no evidence of structural failure
or rejection by the recipient.
dTake Home Message: The human acellular vessel is a
promising experimental therapy, and early clinical
experience supports the continued development
this product for arterial bypass and reconstruction.
2
Gutowski et al
Journal of Vascular Surgery
--- 2020
This study was conducted in full conformity with the
Declaration of Helsinki (revised 2008), Good Clinical Prac-
tice, and the International Council for Harmonisation re-
quirements for Good Clinical Practice. The independent
ethics committee of each participating clinical center
approved the protocol, and each patient provided writ-
ten informed consent before enrolment.
Statistical methods. The primary efficacy end points (pri-
mary, primary assisted,and secondary patency rates of HAV
at month 24) were summarized using Clopper-Pearson
two-sided 95% confidence intervals for binomial pro-
portions.Kaplan-Meier analyseswere used to evaluate time
to loss of patency. The rate and type of graft interventions
and other efficacy variables were analyzed descriptively. All
safety analyses were descriptive.
Investigational product. The Humacyte HAV is a tissue-
engineered HAV composed of human collagen types I
and III and other extracellular matrix proteins, including
fibronectin and vitronectin.
Procedure. Enrolled patients were implanted with a
HAV in the extremity in which they suffered from symp-
tomatic PAD. The proximal and distal target arteries
were surgically exposed in the standard fashion and
the HAV delivered between the two incisions using a
sheath tunneler. Proximal and distal anastomoses were
fashioned in an end-to-side configuration using 5-0 or
6-0 polypropylene sutures (Fig 2). At the conclusion of
implantation, HAV patency and distal arterial runoff to
the foot were confirmed with a high-quality duplex ul-
trasound examination or with conventional intra-
operative angiography (Fig 3).
All patients received 2 days of antibiotic prophylaxis
started intravenously before surgery and continued intra-
venously or intramuscularly. Additionally, antithrombotic
prophylaxis with unfractionated heparin up to 5000 IU
was given intraoperatively and followed by low-
molecular-weight heparin at a prophylactic dose, daily,
until patients were fully mobilized. Starting on the day af-
ter the discontinuation of low-molecular-weight heparin,
dual antiplatelet therapy (aspirin 75-300 mg and clopi-
dogrel 75 mg) was initiated and continued until HAV
abandonment.
HAV interventions and adverse events were recorded at
scheduled all study visits at days 5 and 15; weeks 6, 12,
and 26; months 12, 18, and 24. The patency of the HAV
was determined based on ultrasound findings, graft
Fig 1. Production of human acellular vessels (HAVs). Donor smooth muscle cells are seeded onto a biodegradable
scaffold within a single-use bioreactor. During culture, a cellular bioengineered vessel is grown, which is then
decellularized to produce the HAV.
Journal of Vascular Surgery
Gutowski et al
3
Volume -, Number -
interventions, and clinical examination of the graft site
starting at day 1 and continuing through all follow-up
visits. PAD symptoms, including claudication distance,
pain at rest, and ischemic ulcers, were documented.
Resting ankle to brachial indices (ABIs) were assessed
at baseline and at follow-up visits between 6 weeks
and 24 months. Panel-reactive antibody (PRA; antibodies
directed against class I or II human leukocyte antigens)
and anti-HAV serum IgG levels at 6 months after HAV im-
plantation were compared with those at baseline. The
first vessel in this study was implanted on October 11,
2013, and the final study visit of the last patient enrolled
in the study occurred on May 30, 2016.
Immunoassays. PRAs directed against MHC class I and
class II antigens were assessed in patient serum samples
preoperatively, and at 6 months after implantation, using
clinical-grade testing regimens. Anti-HAV IgG antibodies
were also assessed at these two time points. To measure
levels of anti-HAV IgG antibodies, a customized enzyme-
linked immunosorbent assay was used. HAV samples
were coated onto multiwell plates, and patient serum
samples were incubated. After washing, a monoclonal
antibody directed against human IgG was used in a
sandwich-type enzyme-linked immunosorbent assay
reaction with a colorimetric readout. Pooled human sera
not exposed to the HAV were used as a negative control.
Explant histology. In this study, explanted HAV sam-
ples were obtained from three patients during an other-
wise indicated surgical revision of the HAV. Explanted
HAV tissues were fixed in 10% neutral buffered formalin,
embedded into paraffin, and sectioned (5
m
m thick) for
staining. After deparaffinization and rehydration, routine
hematoxylin and eosin (Statlab reagents; Statlab, McKin-
ney, Tex) staining was performed. Immunofluorescence
staining was performed using a protocol similar to that
previously described.
15
Primary antibodies for human
CNN1 (1:50, ab700; Abcam, Cambridge, UK), alpha
smooth muscle actin (1:200, Abcam, ab5694), CD31 (1:300,
Abcam, ab32457), and CD34 (1:50, #3569; Cell Signaling
Technologies, Danvers, Mass) were incubated on tissue
sections overnight at 4C. Fluorescence conjugated sec-
ondary antibodies (A11001 and A11012; Invitrogen, Carls-
bad, Calif) were diluted 1:400 and incubated on the
tissue sections for 1 hour at room temperature. Slides
were mounted with medium containing 4,6-diamidino-
2-phenylindole to counterstain the nuclei. Sections were
imaged using a Nikon TE2000U microscope equipped
with a Photometrics CoolSNAP HQ2 camera (Tucson,
Ariz). Image acquisition and processing was done using
m
Manager and Fiji software (Image J software).
RESULTS
A single HAV was implanted into 20 patients over the
course of a study enrolment period 7 months. All pa-
tients were Caucasian and 65% were male, with a me-
dian age of 66 years. Demographics are summarized in
Table I. Overall, 75% of patients had hypertension, 45%
had diabetes, and 25% were current tobacco smokers;
additional comorbidities are also summarized in Table I.
The proximal anastomosis was fashioned to the com-
mon femoral artery in 17 patients and to the proximal
SFA in three patients; the distal anastomosis was fash-
ioned to the distal SFA in 11 patients and to the proximal
popliteal artery in nine patients. The median length of
arterial occlusion was 20.2 cm (range, 10.1-28.0 cm) as
assessed by computed tomography or conventional
angiography. All patients had a total segmental occlu-
sion of the SFA with five patients having type B and 15 pa-
tients having type C lesions under the Inter-Society
Consensus for the Management of Peripheral Arterial
Disease II classification system.
14
None involved occlusion
of the popliteal artery. The median total HAV length used
for the bypass procedure was 28 cm (range, 23-30 cm).
The mean follow-up was 20.7 months with a cumulative
follow-up of 34.4 patient-years.
Fig 2. Human acellular vessel (HAV) implantation. Proximal (A) and distal (B) anastomoses of the first HAV
implant for arterial bypass performed in the left lower extremity on October 11, 2013 (Szczecin, Poland). Photos
courtesy of Humacyte, Inc.
4
Gutowski et al
Journal of Vascular Surgery
--- 2020
Seven patients discontinued the study before the
2-year end point; four after graft occlusion and three
who died before study completion. No patients were
lost to follow-up. For the three reported deaths, HAVs
were known to be functional at the patient’s last study
visit. There were no HAV-related deaths. One death was
due to cardiopulmonary failure approximately 2 weeks
after implantation, one was due to metastatic small cell
lung cancer 22 months after implantation, and one
death occurred at 19 months after implantation implant
Fig 3. Arteriograms. Preoperative arteriogram revealing occlusion of the right superficial femoral artery (SFA), but
with suitable proximal target (common femoral artery) and distal target (proximal popliteal artery) for above knee
bypass (A). Arteriogram 5 months after implantation with patent HAV bypass (B).
Journal of Vascular Surgery
Gutowski et al
5
Volume -, Number -
owing to an undetermined cause. None of the events
were considered related to the investigational product
or the surgical procedure.
During the study, two patients developed rest pain at
the 12 months follow-up visit, which resolved after suc-
cessful intervention on the HAV. One patient reported
rest pain and ischemic ulcers at 18 months follow-up
visit. These resolved after replacement of the HAV with
surgical revision to a more distal site using a synthetic
graft.
Patency and interventions. With deaths censored,
patency probability rates obtained by Kaplan Meier esti-
mates of 24-month primary, primary assisted, and sec-
ondary patency rates were 58%, 58%, and 74%,
respectively (Table II;Fig 4). Six of the 20 patients (30%)
required at least one intervention on the HAV during the
course of the study and in total, nine interventions were
performed over the course of the study. These in-
terventions were either a thrombectomy (n ¼4), an an-
gioplasty procedure (n ¼4), or thrombectomy/
angioplasty performed concurrently (n ¼1). Most in-
terventions were successful at restoring patency. How-
ever, in one patient, the graft patency could not be
restored and the HAV was replaced with synthetic
bypass graft material to a more distal target. Two pa-
tients who had previously undergone successful in-
terventions developed a recurrent thrombosis which was
not treated and the HAV was left occluded. Two patients
experienced HAV thrombosis with no or minimal symp-
toms and refused interventions on the HAV.
Ultrasound examinations of the HAVs were performed
at days 5 and 15, and at weeks 6, 12, 26, and at months
12, 18, and 24 (Table III). Midvessel inner diameters were
recorded for each vessels examined and at each time
point. The average midvessel diameter began at
6.0 mm (range, 5.6-6.1 mm) at day 5 and decreased to
5.2 mm by month 24 (range, 2.9-5.8 mm). The largest
midvessel diameter recorded was 7.0 mm. These data
show that the HAVs were mechanically stable during
the follow-up period and did not develop aneurysmal
dilatation in any patient (aneurysmal dilatation being
defined as >50% increase in diameter, or 9 mm).
The median claudication distance was 50 m at
Screening. Six weeks after implantation of the HAV, me-
dian claudication distance increased to 1000 m and
remained at this level for all patients with a patent
HAV. The median resting ABI at screening was 0.64
and, by week 6 post HAV implantation, the median ABI
had increased to normal (1.0); at months 3, 12, and 24,
the median ABI was reported as 1.0, 0.9, and 0.96, respec-
tively (Fig 5). None of the patients with a patent HAV
experienced a decrease in resting ABI.
Complications and infection. Twenty-six HAV-related
complications were reported in 10 patients and
included thrombosis, anastomotic stenosis, HAV stenosis,
pseudoaneurysm, and postprocedural hematoma. The
most frequently reported events were HAV thrombosis (7
patients) and anastomotic stenosis (4 patients). One pa-
tient with multiple HAV thromboses had rest pain at
screening and month 12 follow-up visit. There were a
total of 25 procedural (index surgery) related events
including lymphocele, local swelling, wound infection,
and seroma. There were no HAV-related infections re-
ported during the study and no amputations of the
treated extremity were reported. Three surgical site in-
fections were reported; two were superficial wound in-
fections and one infection that developed in association
with a postoperative lymphocele. The most common
events are listed in Table IV.
One patient developed a pseudoaneurysm that was
noted at the 3-month study visit. The pseudoaneurysmal
segment was excised and replaced with an interposition
segment of expanded polytetrafluoroethylene and the
bypass remained patent through the duration of the
study. The histologic assessment of this specimen
revealed that the HAV had vascular cells within the
wall, indicating signs of remodeling with appropriate
cell types. A hole in the vessel was apparent at the site
of the pseudoaneurysm on both gross and histologic ex-
amination. Overall, the histologic assessment showed lit-
tle to no inflammation, no infection, and no sign of
immunologic reaction to the graft. All clinical records
that were reviewed suggest that the most likely cause
of the resulting pseudoaneurysm was from iatrogenic
trauma to the vessel during the passage of an
Table I. Patient demographics and medical history
Variable
Demographics
Male 13 (65)
Female 7 (35)
Age, years 66 (54-79)
Body height, cm 167.5 (154-178)
Body weight, kg 79 (60-99)
Body mass index, kg/m
2
28.0 (22.8-34.3)
Caucasian ethnicity 20 (100)
Medical history
Hypertension 15 (75)
Diabetes
a
9 (45)
Current tobacco user 5 (25)
Myocardial infarction 5 (25)
Coronary artery disease 4 (20)
Hyperlipidemia 4 (20)
Arteriosclerosis 3 (15)
Coronary angioplasty 3 (15)
Carotid artery stenosis or restenosis 2 (10)
Values are number (%) or median (range).
a
Includes type 1 diabetes mellitus and type 2 diabetes mellitus.
6
Gutowski et al
Journal of Vascular Surgery
--- 2020
embolectomy catheter in a prior procedure. At the last
study visit at 24 months, the HAV remained patent and
functional.
Immunologic assessment. There were no increases in
PRA class I or class II antibody levels observed from base-
line to week 26 postoperatively. Fifteen patients had no
PRA class I antibodies detected at either baseline or
week 26 time points. Of the remaining five patients
who had class I PRAs at baseline, there was no clinically
significant increase observed for any patient (mean class
I PRA of 16% preoperatively; 18% at week 26;
P
¼.24 by
Student’s paired
t
-test). All class II PRA values were zero
at both time points. There were no significant increases
in anti-HAV IgG levels from baseline in 18 of the 20
patients.
Two patients had increases in anti-HAV IgG levels from
baseline that were more than two-fold greater than the
baseline values and were considered possibly signifi-
cant. The first patient experienced no serious adverse
events and, at month 18, the HAV had maintained pri-
mary patency and the luminal diameter had not
changed from its implantation diameter (6 mm). The
patient ultimately died of metastatic lung cancer
before the 24-month visit, and the HAV had remained
patent at the time of death.
The second patient was hospitalized 6 weeks after HAV
implantation with an infected lymphocele in the surgical
Table II. Kaplan-Meier analysis: Patency rates
Patency
Month 6 Month 12 Month 18 Month 24
No.
Probability,
% 95% CI No. Probability, % 95% CI No. Probability, % 95% CI No. Probability, % 95% CI
Primary 15 79 54-92 12 63 38-81 12 63 38-81 4
a
58 33-76
Primary
assisted
15 79 53-92 12 63 38-80 12 63 38-80 4
b
58 33-76
Secondary 17 90 64-97 16 84 59-95 15 79 53-92 6
c
74 48-88
CI,
Confidence interval for graft patency probability;
No.,
number of patients still at risk.
For months 6, 12, 18, and 24, days 180, 360, 540, and 720 were used (see Fig 4), although the visits of the patients may have taken place at different
actual days owing to allowed time windows. Patients who died with the graft still patent were censored at that time point.
a
Five patients completed the study after the last loss of primary patency event at day 559 and before day 720, and were not at risk at day 720.
b
Five patients completed the study after the last loss of primary assisted patency event on day 561 and before day 720, and were not at risk at day 720.
c
Six patients completed the study after the last loss of secondary patency event on day 561 and before day 720, and were not at risk at day 720.
Fig 4. Kaplan-Meier plot. Primary (
blue
) and Secondary (
red
) patency overlay. Patients at risk shown for the
following key time points: days 30, 90, 180, 270, 360, 540, and 720 (ie, months 1, 3, 6, 9, 12, 18, and 24). Patients who
discontinued early or who died with the graft still patent were censored at that time point.
Journal of Vascular Surgery
Gutowski et al
7
Volume -, Number -
wound bed, located close to the distal graft anastomosis.
The lymphocele was not in contact with the HAV anasto-
mosis. The lymphocele was evacuated, and, at 8 weeks
after implantation, the HAV was patent and the wound
was healing. The HAV then occluded sometime between
the 11- and the 24-week postimplantation visits without
recurrence of ischemic symptoms in the treated leg;
therefore, no intervention was done to restore patency.
There was no evidence of HAV dilatation on any ultra-
sound studies and no systemic or local inflammatory
response was described.
In vivo vascular remodeling. No infections of the HAV
vessel itself were reported during this study. This finding
is consistent with the low infection rate observed when
the HAV is used for dialysis access, suggesting that the
remodeling of the HAV matrix by the recipient patient’s
own cells may confer resistance to infections, which can
be problematic with synthetic grafts.
10
In this study,
biopsies of the HAV were obtained from three patients
at 13, 50, and 61 weeks after implantation. In all cases,
samples of HAV were obtained during a surgical pro-
cedure that was clinically indicated for HAV revision or
repair. Hematoxylin and eosin staining of explant sam-
ples (Fig 6,
A1
,
B1
,and
C1
)showprogressiveinfiltration of
the HAV wall with spindle-shaped cells (Fig 6). Immu-
nostaining for smooth muscle markers shows yellow
co-staining for smooth muscle alpha actin (red) and
calponin (green) in the HAV wall, indicating a smooth
muscle phenotype of repopulated cells (Fig 6,
A2
,
B2
,
and
C2
). Interestingly, the appearance of CD34
þ
cells in
the neoadventitia at the early 13-week time point is
accompanied by multiple CD31
þ
microvessel structures,
perhaps implying recruitment of CD34
þ
vascular
endothelial progenitors to the outer surface of the HAV
(Fig 6,
A3
and
B3
). Over time, it appears that the CD34
þ
cells become less frequent, and instead the appearance
of robust CD31
þ
microvessels in the outer HAV media
are evident at 61 weeks (Fig 6,
C3
). Although the luminal
surfaces of the HAV biopsies did not seem to stain for
endothelial markers (Fig 6,
A3
,
B3
,and
C3
insets), given
that these vessels typically underwent intravascular
manipulations/ballooning, it is not surprising that any
luminal endothelium might have been stripped.
DISCUSSION
The optimal conduit for peripheral arterial bypass has
yet to be established
16
and typical processes to use autol-
ogous venous conduit are associated with morbidity,
increased health care use, and costs secondary to wound
complications.
17-19
Wound complications and infection
from venous harvest sites occur in as many as 25% of
cases using autologous vein.
17,18
Synthetic alternatives
provide additional options when autologous conduit is
not available, but these grafts pose a disadvantage to
host tissue in terms of patency, biocompatibility, infec-
tion, and durability.
20,21
The increasing incidence of
atherosclerotic vascular disease and the short-comings
of native autologous vein and synthetic materials only
highlight the need for a better vessel replacement
option.
The HAV, when used as a conduit for dialysis access,
withstands repeated cannulation over periods of more
than 1 year without aneurysm or structural degradation
and does not require a prolonged time for maturation.
10
There is a low infection rate observed when the HAV was
used as a conduit for dialysis access or a vascular bypass
for patients with PAD. The potential advantages of the
HAV may include lower complication rates and
decreased infections, leading to better long-term graft
survival. Early clinical observations in this study require
additional testing in larger, prospective clinical trials.
The overall goal of regenerative therapies is to repair or
replace damaged tissue with new therapies that biolog-
ically mimic the failing tissue. Our prior report on the use
of human bioengineered blood vessels for hemodialysis
access demonstrated the feasibility and the potential
benefits of using regenerative therapies for vascular
replacement.
10
In that study, 60 HAVs were cannulated
with large-bore dialysis needles three times per week.
There were no negative immune reactions to the HAV,
no reports of vessel degeneration or unexpected conduit
failures, a low incidence of overall infection, and favor-
able enduring patency. Moreover, HAV tissue samples
explanted from that study population revealed substan-
tial host recellularization that transformed the once acel-
lular HAV into the patient’s own living blood vessel.
10
A
bioengineered arterial replacement, like the HAV, that
behaves like native tissue within the host could address
an unmet medical need in the future. This pilot study
of 20 patients is the first-in-man experience with a hu-
man bioengineered blood vessel as a conduit for periph-
eral arterial bypass.
In this trial and in our previously reported hemodialysis
access trial, there was no clinical, ultrasound, or angio-
graphic evidence of structural degradation or true
aneurysm formation. In the patient in this study who
Table III. Midvessel diameter (mm) assessed by ultra-
sound examination
Visits No. Median Range
Day 5 20 6.0 5.6-6.1
Day 15 20 6.0 5.2-7.0
Week 6 19 5.8 4.5-6.1
Week 12 18 5.9 4.6-6.4
Week 26 18 5.7 4.6-6.7
Month 12 15 5.4 2.7-6.8
Month 18 15 5.5 2.5-6.6
Month 24 13 5.2 2.9-5.8
No.,
Number of patients with observations.
8
Gutowski et al
Journal of Vascular Surgery
--- 2020
developed the pseudoaneurysm, clinical and histologic
information suggests that the initial defect was most
likely caused by iatrogenic injury during the early postop-
erative period.
There was no evidence of immune rejection of the HAV,
as detected from clinical explants or from the 6-month
serologic assessments. There was no increase in PRA
class I or class II levels observed from baseline to week
26. Two patients experienced an increase in anti-HAV
IgG that was not associated with any adverse clinical
events, such as dilatation or aneurysm formation, nor
did this seem to alter the structural integrity of the
HAV during the study period. In contrast, cryopreserved
human veins are subject to immunologic complications,
and these reactions can result in structural instability and
aneurysmal degeneration.
6
Tissue samples were obtained from implanted HAVs of
three patients; specimens were collected at the time of
surgical reexposure for open thrombectomy or technical
revisions at 13, 50, and 61 weeks in three different pa-
tients. Immunostaining techniques revealed positive
markers for smooth muscle cells (eg, smooth muscle
actin/calponin) and endothelial cells (CD31), and
together, these markers are suggestive of positive
vascular remodeling. Grossly, there is an appearance
that over time, the HAV structurally develops into a vessel
with arterial attributes as is demonstrated in Fig 7, where
the medial wall of the vessel demonstrates bleeding
when transected (Supplementary Video, online only).
Primary patency was 63% at 12 months and 58% at
24 months, and secondary patency was 84% at 12 months
and 74% at 24 months. The initial results from this small
phase II study reflect positively on the potential for the
HAV to provide long-term bypass durability.
There were no cases of direct HAV infection in 34.4
patient-years of follow-up; three surgical site infections
were reported (two postoperative wound infections and
one infected lymphocele). In the patient with the
infected lymphocele, a lymphatic cyst developed at the
surgical exposure site (close to the distal graft anasto-
mosis) that subsequently became infected and required
patient hospitalization, surgical drainage, and antibiotics.
However, neither the infected lymphocele nor the post-
operative wound infections progressed to subsequent
HAV infection. Similar to our previous trial in hemodialy-
sis access, there was a very low rate of overall infection in
this study (0.09 per patient-year, or three cases in 34.4
Fig 5. Ankle-brachial indices (
ABIs
) over time. The ABI values for each patient in the study at the following key
time points: weeks 6, 12, and 26 and months 12, 18, and 24. Patient 2, 4, 12, 14, 15, and 16 had graft complications and
patient 20 died, with no data recorded.
Table IV. Human acellular vessel (
HAV
) complications and
surgical site infections
Complication/infection
All patients
(N ¼20)
No. of
events
No. (%) of
patients
Total HAV
complications
26 10 (50)
HAV thrombosis 15 7 (35)
Anastomotic stenosis 6 4 (20)
HAV stenosis 2 2 (10)
HAV pseudoaneurysm 2 1 (5)
Postprocedural hematoma 1 1 (5)
Total surgical site infection 3 3 (15)
Postoperative wound infection 2 2 (10)
Infected lymphocele 1 1 (5)
Journal of Vascular Surgery
Gutowski et al
9
Volume -, Number -
patient-years), none of which led to the abandonment or
explantation of the HAV.
Rehospitalization for wound complications after pe-
ripheral bypass is common, and occurs approximately
10% to 20% of the time, leading to increased morbidity,
mortality, limb loss, patient dissatisfaction, and cost to
the health care system.
22-24
Rates of wound complica-
tions and infection after lower extremity bypass are
similar between prosthetic and autologous conduit and
range between 4.5% and 6.5%.
20,22
Lymphocele forma-
tion and other wound complications after lower extrem-
ity bypass are common in this patient population.
Fig 6. Histologic evaluation of explanted human acellular vessels (HAVs) at 13 (A),50(B), and 61 (C) weeks after
implantation. Sections of explanted HAV tissue samples stained with hematoxylin and eosin (A1, B1, C1) show
development of neoadventitial (a) and medial (m) layers. High magnification inset taken from region denoted
with an asterisk (*). The anastomotic suture hole in B1 is identified with black arrow. Immunofluorescence staining
of alpha smooth muscle actin (
a
SMA,
red
) and calponin (CNN1,
green
) reveal myogenic host recellularization (A2,
B2, C2). Coexpression (
yellow overlay
) of CNN1 with
a
SMA indicative of maturation of smooth muscle cells.
Microvasculature within neoadventitia express endothelial marker CD31 (
red
;A3, B3, C3), and expression of the
early endothelial marker CD34 (
green
) is predominantly observed at 13 weeks after implantation (A3). All three
explanted samples lacked substantial presence of CD31
þ
endothelial cells on lumen (A4, B4, C4). Nuclei (
blue
)
were counterstained with 4,6-diamidino-2-phenylindole.
10
Gutowski et al
Journal of Vascular Surgery
--- 2020
Because this was a phase II clinical trial, the study was
limited by the small sample size, and it was not
adequately powered to assess efficacy. Furthermore,
the study population was entirely European, Caucasian,
and predominantly male. Despite these limitations, we
have demonstrated the feasibility of using a bio-
engineered vessel in a high pressure arterial circuit to
treat PAD. There were no evidence of vessel rejection or
spontaneous vessel degradation, and the HAV func-
tioned as intended as a vascular bypass conduit.
CONCLUSIONS
In this phase II study, the HAV functioned as intended
as a peripheral arterial bypass conduit. The vessel suc-
cessfully restored in-line blood flow to the distal extrem-
ity and foot, and in doing so led to objective and
subjective improvements in patients suffering from
chronic limb ischemia secondary to PAD. All patients
shared the initial benefit of symptomatic relief from
rest pain and claudication and the majority of the cohort
sustained improvement in ABI from baseline. There were
no amputations (minor or major) of the index limb dur-
ing the 24-month study period. There is histologic evi-
dence that the HAV remodels with host cells as early as
3 months. Patency rates are within the ranges of patency
rates of synthetic and autologous venous grafts pre-
sented in the literature. The results support the
continued development of human tissue engineered
blood vessels for vascular reconstruction.
AUTHOR CONTRIBUTIONS
Conception and design: SG, RK, LN, AP, HP, BT, JL
Analysis and interpretation: SG, RK, LN, AP, HP, BT, JL
Data collection: PG, MG, MI, SP, RS, JT, WW, NZ, TZ
Writing the article: PG, SG, MG, MI, AK, RK, LN, AP, HP, SP,
RS, BT, JT, WW, NZ, TZ, JL
Critical revision of the article: PG, SG, MG, MI, AK, RK, LN,
AP, HP, SP, RS, BT, JT, WW, NZ, TZ, JL
Final approval of the article: PG, SG, MG, MI, AK, RK, LN,
AP, HP, SP, RS, BT, JT, WW, NZ, TZ, JL
Statistical analysis: Not applicable
Obtained funding: Not applicable
Overall responsibility: JL
REFERENCES
1. Rob CG. A history of arterial surgery. Arch Surg 1972;105:821-3.
2. Kunlin J. Long vein transplantation in treatment of ischemia
caused by arteritis. Rev Chir 1951;70:206-35.
3. Julian OC, Dye WS, Olwin JH, Jordan PH. Direct surgery of
arteriosclerosis. Ann Surg 1952;136:459.
4. Almasri J, Adusumalli J, Asi N, Lakis S, Alsawas M, Prokop LJ,
et al. A systematic review and meta-analysis of revasculari-
zation outcomes of infrainguinal chronic limb-threatening
ischemia. J Vasc Surg 2018;68:624-33.
5. Holdsworth RJ, Naidu S, Gervaz P, McCollum PT. Glutaral-
dehyde-tanned bovine carotid artery graft for infrainguinal
vascular reconstruction: 5-year follow-up. Eur J Vasc Endo-
vasc Surg 1997;14:208-11.
6. Carpenter JP, Tomaszewski JE. Human saphenous vein
allograft bypass grafts: immune response. J Vasc Surg
1998;27:492-9.
7. Scharn DM, DirvenM, Barendregt WB, Boll APM, Roelofs D,
van der Vliet JA. Human umbilical vein versus
Fig 7. Intraoperative photo. Taken during open surgical
intervention 12 months after the initial implant. The hu-
man acellular vessel (HAV) has developed appearance of
an intima (
yellow arrow
), media (
green arrow
), and
adventitia (
blue arrow
). Bleeding from the HAV edges
(
white arrows
) serves as evidence that the HAV has un-
dergone neovascularization and developed a microvas-
cular network. (Images curtesy of Jakub Turek and Norbert
Zapotoczny, Wroclaw, Poland.)
Journal of Vascular Surgery
Gutowski et al
11
Volume -, Number -
heparin-bonded polyester for femoro-popliteal bypass:
5-year results of a prospective randomized multicenter trial.
Eur J Vasc Endovasc Surg 2008;35:61-7.
8. Dahl SL, Koh J, Prabhakar V, Niklason LE. Decellularized
native and engineered arterial scaffolds for transplantation.
Cell Transplant 2003;12:659-66.
9. Dahl SL, Kypson AP, Lawson JH, Blum JL, Strader JT, Li Y,
et al. Readily available tissue-engineered vascular grafts. Sci
Transl Med 2011;3:68ra9.
10. Lawson JH, Glickman MH, Ilzecki M, Jakimowicz T,
Jaroszynski A, Peden EK, et al. Bioengineered human acel-
lular vessels for dialysis access in patients with end-stage
renal disease: two phase 2 single-arm trials. Lancet
2016;387:2026-34.
11. Niklason LE, Gao J, Abbott WM, Hirschi KK, Houser S,
Marini R, et al. Functional arteries grown in vitro. Science
1999;284:489-93.
12. Poh M, Boyer M, Solan A, Dahl SL, Pedrotty D, Banik SS, et al.
Blood vessels engineered from human cells. Lancet
2005;365:2122-4.
13. Niklason LE, Abbott W, Gao J, Klagges B, Hirschi KK,
Ulubayram K, et al. Morphologic and mechanical char-
acteristics of engineered bovine arteries. J Vasc Surg
2001;33:628-38.
14. Hardma RL, Jazaeri O, Yi J, Smith M, Gupta R. Overview of
classification systems in peripheral artery disease. Semin
Intervent Radiol 2014;31:378-88.
15. Kajimura J, Ito R, Manley NR, Hale LP. Optimization of single-
and dual-color immunofluorescence protocols for formalin-
fixed, paraffin-embedded archival tissues. J Histochem
Cytochem 2016;64:112-24.
16. Conte MS. The ideal small arterial substitute: a search for the
Holy Grail? FASEB J 1998;12:43-5.
17. Santo VJ, Dargon PT, Azarbal AF, Liem TK, Mitchell EL,
Moneta GL, et al. Open versus endoscopic great saphenous
vein harvest for lower extremity revascularization of critical
limb ischemia. J Vasc Surg 2014;59:427-34.
18. Bitondo JM, Daggett WM, Torchiana DF, Akins CW,
Hilgenberg AD, Vlahakes GJ, et al. Endoscopic versus open
saphenous vein harvest: a comparison of postoperative
wound complications. Ann Thorac Surg 2002;73:523-8.
19. Luckraz H, Kaur P, Bhabra M, Mishra PK, Nagarajan K,
Kumari N, et al. Endoscopic vein harvest in patients at high
risk for leg wound complications: a costebenefit analysis of
an initial experience. Am J Infect Control 2016;44:1606-10.
20. Klinkert P, Post PN, Breslau PJ, Van Bockel JH. Saphe-
nous vein versus PTFE for above-knee femoropopliteal
bypass. A review of the literature. Eur J Vasc Endovasc
Surg 2004;27:357-62.
21. van de Weijer MA, Kruse RR, Schamp K, Zeebregts CJ,
Reijnen MM. Morbidity of femoropopliteal bypass surgery.
Semin Vasc Surg 2015;28:112-21.
22. Davis FM, Sutzko DC, Grey SF, Mansour MA, Jain KM,
Nypaver TJ, et al. Predictors of surgical site infection after
open lower extremity revascularization. J Vasc Surg 2017;65:
1769-78.
23. Peacock MR, Shah NK, Farber A, Lee SY, Kalish JA, Rybin D,
et al. Index complications predict secondary complications
after infrainguinal lower extremity bypass for critical limb
ischemia. J Vasc Surg 2017;65:1344-53.
24. McPhee JT, Barshes NR, Ho KJ, Madenci A, Ozaki CK,
Nguyen LL, et al. Predictive factors of 30-day unplanned
readmission after lower extremity bypass. J Vasc Surg
2013;57:955-62.
Submitted May 29, 2019; accepted Nov 23, 2019.
Additional material for this article may be found online
at www.jvascsurg.org.
DISCUSSION
Dr Ahmad Alsheekh (
Brooklyn, NY
). I would like to
thank you for this interesting study and I would like to
ask about these grafts. Is it tolerable to post-implantation
interventions like stenting and ballooning after implant-
ing these grafts?
Dr Jeffrey H. Lawson. So, we now have a real-world
experience in a little more than 200 human acellular ves-
sels implanted. These function as human vascular tissues,
so they’ve now been put through the paces of regular
clinical care. Initially, we babied them and didn’tdoa
lot, but now they’ve been angioplastied, they’ve been
stented, they’ve been revised, and they can tolerate all
that. They can tolerate thrombectomy. I did the initial
thrombectomies open and we’ve done now percuta-
neous thrombectomy.
The only thing we’ve learned is, which is a little atypical,
you know, we tend to oversize balloons in a lot of inter-
ventions. In this case, if you oversize a balloon, if it’sa
6mm conduit, you can usually go 6 mm for an angio-
plasty in the conduit itself, 7 mm is okay at the anasto-
mosis; but if you go bigger, you can actually disrupt the
vessel. So we’ve had a couple of vessels that were
terminally disrupted by basically a 9 mm balloon that
can rip the vessel.
12
Gutowski et al
Journal of Vascular Surgery
--- 2020
