Cytotherapy 26 (2024) 1076 1083

Contents lists available at ScienceDirect

CYTOTHERAPY journal homepage: www.isct-cytotherapy.org

FULL-LENGTH ARTICLE
Stromal Cell Therapy

Adipose-derived stromal vascular fraction cells to treat long-term

pulmonary sequelae of coronavirus disease 2019: 12-month follow-up
Michael Carstens1,2,*, Jessy Trujillo3, Yanury Dolmus4, Carlos Rivera5, Santos Calderwood6,

pez7, Kenneth Bertram2
Judith Lejarza1, Carlos Lo
1
Department of Surgery, Hospital Escuela Oscar Danilo Rosale Argu€ ello, Leo
n, Nicaragua
2
Wake Forest Institute for Regenerative Medicine, Winston-Salem, North Carolina, USA
3
Department of Medicine, Hospital Monte Espan ~ a, Managua, Nicaragua
4
Department of Pediatrics, Hospital Escuela Cesar Amador Molina, Matagalpa, Nicaragua
5
Department of Radiology, Hospital Escuela Cesar Amador Molina, Matagalpa, Nicaragua
6
Department of Surgery, Hospital Escuela Cesar Amador Molina, Matagalpa, Nicaragua
7
Department of Medicine, Hospital Escuela Oscar Danilo Rosales Argu € ello, Leo

n, Nicaragua

A R T I C L E I N F O A B S T R A C T

Article History: Background aims: Long coronavirus disease (COVID) is estimated to occur in up to 20% of patients with corona-
Received 6 February 2024 virus disease 2019 (COVID-19) infections, with many having persistent pulmonary symptoms. Mesenchymal
Accepted 15 March 2024 stromal cells (MSCs) have been shown to have powerful immunomodulatory and anti-fibrotic properties. Autol-
ogous adipose-derived (AD) stromal vascular fraction (SVF) contains MSC and other healing cell components
Key Words: and can be obtained by small-volume lipoaspiration and administered on the same day. This study was
adipose-derived mesenchymal cells
designed to study the safety of AD SVF infused intravenously to treat the pulmonary symptoms of long COVID.
DLCO
Methods: Five subjects with persistent cough and dyspnea after hospitalization and subsequent discharge for
long COVID
pulmonary fibrosis
COVID-19 pneumonia were treated with 40 million intravenous autologous AD SVF cells and followed for
stromal vascular fraction 12 months, to include with pulmonary function tests and computed tomography scans of the lung.
Results: SVF infusion was safe, with no significant adverse events related to the infusion out to 12 months.
Four subjects had improvements in pulmonary symptoms, pulmonary function tests, and computed tomog-
raphy scans, with some improvement noted as soon as 1 month after SVF treatment.
Conclusions: It is not possible to distinguish between naturally occurring improvement or improvement
caused by SVF treatment in this small, uncontrolled study. However, the results support further study of
autologous AD SVF as a treatment for long COVID.
© 2024 International Society for Cell & Gene Therapy. Published by Elsevier Inc. This is an open access article
under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

Introduction 3 months from the onset of COVID-19 with symptoms that last for at
least 2 months and cannot be explained by an alternative diagnosis”
Since the emergence of coronavirus disease 2019 (COVID-19), the [2]. WHO reports that approximately 10 20% of people infected with
World Health Organization (WHO) estimates that there have 774 000 COVID-19 may go on to develop symptoms that can be diagnosed as
000 COVID-19 cases and 7 000 000 deaths as of mid-January 2024 long COVID. There are currently no validated effective treatments.
[1]. In addition, many patients have continued to have symptoms The total global cost of long COVID (health care, lost productivity and
attributed to COVID-19, lasting for months to now years after the pri- decreased quality of life) is estimated at $3.7 trillion (USD) [3].
mary viral infection. WHO has defined “long COVID” as the persis- Long COVID affects multiple organ systems, with common new-
tence of COVID-19 associated symptoms “in individuals with a onset conditions including pulmonary, cardiovascular and cerebro-
history of probable or confirmed SARS-CoV-2 infection, usually vascular diseases. Up to 40% of long COVID patients complain of
shortness of breath, and 20% complain of cough. Possible mechanisms
* Correspondence: Michael Carstens, Wake Forest University School of Medicine, of pulmonary damage may be related to systemic inflammation and/
4230 Foxbury Court, Winston-Salem, NC 27104, USA. or continued presence of COVID-19 virus [4] Autopsy studies of
E-mail addresses: MichaelCarstens@mac.com, mcarsten@wakehealth.edu patients with COVID-19 document a pattern of diffuse alveolar
(M. Carstens).

https://doi.org/10.1016/j.jcyt.2024.03.491
1465-3249/© 2024 International Society for Cell & Gene Therapy. Published by Elsevier Inc. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/)
 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083 1077

damage consistent with two mechanisms: (1) viral attack on the vas- concluded that MSC therapy was safe, with good immune tolerance,
cular endothelium resulting in reduction of the available surface area and significantly reduced lung damage. They further concluded that
for the absorption of oxygen; and (2) immune-mediated activation of the benefit of MSCs in COVID-19 was the result of their immunomo-
fibrosis, leading to deposition of scars in the alveolar walls and inter- dulation, anti-fibrotic and tissue repair properties.
stitium, resulting in a physical barrier for the diffusion of oxygen Adipose tissue provides a readily accessible sources of MSCs; these
[5 8]. Thus, COVID-19 pneumonia can inflict extensive parenchymal cells can be isolated from the adipose structural components after lipo-
damage, leading to long-term scarring and physiologic dysfunction. suction [27 30]. Leng et al. produced clinical-grade AD-MSCs as a cel-
Mo et al. [9] were the first to report the persistence of pulmo- lular product under Good Manufacturing Practice laboratory
nary function abnormalities in 110 patients with COVID-19 at dis- conditions to treat seven patients with COVID-19 pneumonia using
charge, the most common being those of diffusing capacity of the 1 £ 106 cells/kg body weight via intravenous infusion. Their results
lungs for carbon monoxide (DLCO) and total lung capacity (TLC). showed that AD-MSCs were safe, that the patients had better func-
With increasing severity of the initial illness, these deficits are tional outcomes, with improved chest computed tomography (CT)
more pronounced. DLCO was reduced in 47% of all patients and imaging and multiple laboratory studies demonstrating a decrease in
in 84% of patients with severe initial symptoms. Similarly, 25% of the systemic pro-inflammatory state related to COVID-19 cytokine
patients with less-aggressive disease had a TLC <80% predicted; storm [31]. Specifically, pro-inflammatory cells, C-reactive protein and
this rose to 47% of patients with severe pneumonia. These finding tumor necrosis factor-alpha were decreased, whereas regulatory cells
are supported by Thomas et al. [10], who identified DLCO deficits and interleukin (IL)-10 increased. Shetty et al., in an accompanying edi-
in 20 30% of patients with mild-moderate disease and in 60% of torial, highlighted the benefits of intravenous infusion as an efficient
those with severe disease. Lung volumes in survivors were method to deliver AD-MSC to lungs, where paracrine factors produced
assessed by Foti et al. [11] stratified for initial disease, with an by the MSC could help reduce the overactivation of the immune sys-
average volume of 4262 cc in “good outcome” patients compared tem, protect alveolar cells, counteract fibrosis and improve lung func-
with 3578 cc in “bad outcome” patients. tion [32]. With the use of a different dosing regimen, 17 patients with
Bellan et al. [12] studied 238 patients who had COVID-19 infection COVID-19 who required mechanical ventilation received between one
severe enough to require hospitalization. Patients were studied at 4 and three doses of AD-MSC intravenously at 1 £ 106 cells/kg of recipi-
months after hospital discharge and then again at 1 year after dis- ent’s body weight. The study demonstrated safety and overall clinical
charge. At 4 months, dyspnea persisted in approximately 10% of improvement in most patients. In addition, laboratory studies showed
patients who reported dyspnea during their acute phase of COVID-19 a decrease in inflammatory serum markers (C-reactive protein, IL-6,
infection, possibly attributable to pulmonary fibrosis. In addition, ferritin, lactate dehydrogenase and D-dimer) [33]. Another dosing regi-
more than 50% of the patients had a DLCO less than 80% of expected, men (100 million cells £ 4 infusions) used commercially manufactured
indicating abnormal oxygen and carbon dioxide diffusion capacity AD-MSC and demonstrated safety [34]. Gentile et al. reviewed ongoing
across the pulmonary alveoli. Fifteen percent of patients had a DLCO clinical trials using cultured/manufactured AD-MSC to treat severe
less than 60% expected, indicating severe diffusion impairment. At COVID-19 or COVID-19 associated acute respiratory distress syn-
the 12-month follow-up, 49% of patients still had a DLCO less than drome, concluding that AD-MSC are safe and effective [35]. In addition,
80% expected, whereas 10% remained with a DLCO less than 60% they reported in a literature review the role of MSCs, AD-MSCs and adi-
expected. Of the patients with an altered DLCO at 4 months’ post- pocyte-secreted exosomal microRNAs as a potential antiviral therapy,
hospital discharge, 20% showed a significant improvement in DLCO comparing the results found with those related to drugs and vaccines
whereas 29% showed a significant worsening [13]. used in COVID-19 disease [36].
There are, however, disadvantages to the use of AD-MSCs to treat
Mesenchymal stromal/stem cells (MSCs) COVID-19. These include the cost of Good Manufacturing Practice cul-
ture/manufacturing, time of manufacturing and potential allergic
The finding of multipotent stem cells (now called MSCs) derived reactions, as the cell product can be from an allogeneic source. An
from bone marrow that could differentiate in vitro into various cell alternative to cultured/manufactured AD-MSC is an AD-MSC con-
types raised the possibility that these cultured (manufactured) MSCs taining product, stromal vascular fraction (SVF), that can be obtained
could be used to treat multiple diseases [14,15]. MSCs have been and delivered as a point-of-care intervention. Cultured AD-MSCs and
shown to recognize sites of injury, “sense” the environment, and selec- SVFs are two distinct products, both obtained after a liposuction pro-
tively respond through multiple pathways to support the remodeling/ cedure. A volume of 100 mL of harvested adipose tissue would yield
repair of the injured tissues. In particular, the ability of MSC to “down- approximately 100 300 native AD MSCs, which can be multiplied in
regulate” inflammatory pathways presented a new capability/possibil- culture over time (manufactured) to the desired dose of (generally
ity to mitigate inflammation and even heal the underlying tissue dam- allogeneic) homogenous cells. The same volume of harvested tissue,
age [16]. Gregoire et al. [17] evaluated bone marrow MSC after enzymatic treatment, would yield 10 £ 106 autologous SVF cells
administered intravenously to treat severe COVID-19 and found that (which includes the native AD MSC population) and can be delivered
the MSC infusions were well tolerated and showed promising efficacy. back to patient on the same day as the liposuction harvest [37].
MSC cell products have subsequently been isolated and cultured
from other tissues to include placenta, cord blood and adipose tissue Stromal vascular fraction
[18]. Several studies have tested the safety and potential efficacy of
MSC from different sources to treat acute COVID-19 pneumonia [19]. Point-of-care methods to obtain autologous SVFs from adipose tis-
Cultured placenta MSCs were used to treat patients with COVID- sue require isolation and separation of the SVF cells from the adipo-
19 related acute respiratory distress syndrome [20 23]. The pla- cytes and extracellular matrix via tissue processing methods. Using
centa MSCs were safe and improved respiratory distress. Rebelatto an enzymatic processing approach, Kapur et al. and Brown et al.
et al. [24] reported that (allogeneic) umbilical cord MSCs were safe describe a closed-system method used at the point of care to harvest
and decreased the levels of blood pro-inflammatory proteins and the and process cells from the adipose tissue that involves a limited lipo-
extent of lung damage. Shi et al. [25] performed a randomized phase suction procedure (100 cc), followed by a washing process to
2 trial of more than 100 subjects evaluating umbilical cord MSC. One- removed blood cells and debris, treatment with type I collagenase for
year follow-up demonstrated improvement of symptoms and lung collagen digestion, centrifugation and filtration of the cell suspension
lesions with good tolerance. In addition, Shi et al. [26] published a through a 400-micron mesh. The resulting cellular pellet is aspirated,
review of the MSC therapy trials for severe COVID-19. They and the supernatant is discarded. The cell pellet is a heterogenous
1078 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083

mixture of fibroblasts, stromal progenitor cells, endothelial precursor Table 1

cells, tissue-resident macrophages, smooth muscle cells, pericytes Patient demographics.
and preadipocytes [38,39]. The liposuction harvest, SVF isolation and
Subject Age, y Sex Days in the Time (days) from
separation procedures can be conducted in less than 3 hours using a intensive hospital admission to
sterile disposable device, SVF-2 (GID BIO, Louisville, CO, USA) [38]. care unit stromal vascular fraction
SVF and AD MSC are different cell products. However, both prod- treatment
ucts work through secretory (paracrine) and anti-inflammatory activ- 1 46 Female 30 185
ities. Both populations have been shown to decrease inflammation, 2 57 Female 30 257
decrease fibrosis and increase micro-capillary networks [40]. SVF has 3 41 Male 66 125
4 39 Male 11 76
been shown to secrete numerous growth factors and cytokines,
5 32 Male 14 173
including vascular endothelial growth factor, transforming growth Mean 42 30 163
factor-1 and hepatocyte growth factor, and interferon-gamma and
IL-10 [40,41]. In addition, SVF has been shown to induce M2 macro-
phages, further driving an anti-inflammatory environment [42]. male or female, and a body mass index of 22. Pre-enrollment pul-
AD SVF has been used in multiple clinical trials to treat diabetic monary function studies inclusion criteria included forced vital
foot ulcers, peripheral vascular disease, burn wounds, radiation capacity (FVC) 40% predicted and 70% predicted and DLCO 20%
injury and Crohn’s disease fistula, among other disease states predicted and 70% predicted.
[42 45]. SVF has been used in the United States under a Food and Exclusion criteria were as follows: (i) use of home oxygen, (ii) his-
Drug Administration investigational device exemption to treat tory of pulmonary malignancy, (iii) immunosuppressive drug treat-
knee osteoarthritis (safe and with evidence of potential efficacy) ment, (iv) history of previous cardiac disease with an ejection
and is currently in a pivotal clinical trial (»150 subjects) for that fraction of 30%, (v) diabetes, (vi) pregnancy or plans to conceive
indication [46]. during the study period and/or (vii) participation in another clinical
This is the first report of the safety of autologous AD SVF cells study. Although admission to the intensive care unit was not a study
delivered by intravenous infusion to treat pulmonary symptoms in inclusion criterion, it is of note that all five subjects had been admit-
patients with long COVID. In addition, clinical results suggest a poten- ted to the intensive care unit, indicating the severity of their COVID-
tial therapeutic effect from the SVF infusion in this small, non-place- 19 infection.
bo controlled study. Pulmonary function testing was carried out by a single study team
physician at the Pulmonary Laboratory of Hospital Monte Espan ~ a,
Methods Managua using the EasyOne Pro LAB (ndd Medical Technologies
Andover, MA, USA). High-resolution computed tomography was
Study design conducted using a Phillips 128-cut scanner with cuts of 3 mm. CT
scans were scheduled at baseline (pre-SVF infusion), and 3, 6 and
This was a prospective, interventional, open-label, safety study, 12 months after SVF infusion. As the result of CT scheduling issues in
with hypothesis-generating efficacy outcomes, involving the admin- the public hospital, subjects did not get their 3-month CT scans.
istration of a single-dose, 40 £ 106 autologous SVF cells, derived at Therefore, data presented are a comparison between baseline and
point-of-care and delivered via an intravenous infusion. The dose of the 6- and 12-month visits. All CT scans were read by a single study
40 £ 106 SVF cells was based on study team experience with SVF to team radiologist.
treat non-healing diabetic foot ulcers [43]. Safety and efficacy follow-
up visits were carried out at 1, 3, 6 and 12 months. Procedures
This study was conducted within the Ministry of Health (MINSA)
and the Nicaragua national health care system. All clinical procedures
used were in accordance with the ethical standards of the responsible Preparation of SVF
Committee on Human Experimentation (institutional review board SVF isolation and intravenous administration of the SVF cells were
no. NIC-MINSA/CNDR CIRE-05/03/21-112) and the Helsinki Declara- conducted in either the Managua or Leon hospitals. The GID SVF-2
tion of 1975, as revised in 2000. Informed consent was obtained from device (Figure 1) was used to isolate and separate the SVF cells for
all participants included in this case studies. The patients also pro- each patient (GID Bio) [25]. Fifty-eighty cubic centimeters of raw lip-
vided consent for publication. The trial was listed on clinicaltrials.gov oaspirate was harvested from the abdominal subcutaneous fat and
NCT06304623. passed directly into the sterile SVF-2 processing device. The lipoaspi-
Before the trial was started, conversations with MINSA and local rate was washed three times to remove red blood cells, leukocytes
treating physicians about study design indicated that subjects would and free oils. Approximately 125 cc of Hartmann’s solution was
only enroll in this pilot study if they were being “treated.” There was added to the adipose tissue with 70 000 CDU of collagenase enzyme
no support for a control arm at the time of this study. Patients were (Worthington CLS-1, Lakewood, NJ, USA) to achieve a concentration
concerned about being re-exposed to COVID-19 by returning to hospi- of 300 CDU/mL of total volume. The mixture was dissociated for 60
tals (with patients infected with COVID-19) for the study procedures minutes on a rotary mixer in an incubator at 39°C and at 150 RPM.
and were concerned about travel and missing work. The potential eli- After dissociation, the mixture was centrifuged for 10 minutes at
gible patients felt that the risks, missed work and other inconveniences 600g. The SVF cell pellet, located at the bottom of the device, was
were only justified if they were in an active treatment arm. removed with a 14G 6-inch needle. Two SVF samples were passed
through an automated fluorescence cytometer (LUNA-STEM; Logos
Patient selection and pre-SVF treatment evaluations Biosystems, Annandale, VA, USA) for cell counting and viability
assessment. SVF cell viability was 70% in all infusions.
Patients were identified at Managua and Leon, Nicaragua, commu-
nity health centers and screened for eligibility by study physicians Administration of SVF
(Table 1). Five patients previously hospitalized for polymerase chain A volume of the SVF suspension containing a dose of 40 £ 106 SVF
reaction confirmed COVID-19 and with persistent pulmonary com- cells was isolated. The cells were then further suspended to a total of
plaints of dyspnea for at least 2 months after hospital discharge were 100 cc of Hartmann’s solution. The cell suspension was administered
recruited. Additional inclusion criteria included age 18 85 years, intravenously over a period of 30 minutes through a blood filter.
 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083 1079

Fig. 1. SVF-2 tissue-processing device: empty (left) and SVF cell pellet (right). SVF-2 device for production of autologous SVF at point-of-care, used for tissue harvest, washing, disag-
gregation, filtering and centrifuging. Left: device before use. Right: device with SVF pellet in receiving chamber. (Color version of figure is available online.)

Safety monitoring women. The predicted value per patient is determined by height, age
and gender [50,51]. A published MCID value for increased TLC is not
Subjects were followed overnight in the hospital with vital signs available.
and clinical monitoring for any potential (serious) adverse events [(S) DLCO measures the transfer of carbon monoxide from alveolar gas
AE] related to the lipoaspiration procedure (e.g., bleeding) or the SVF to hemoglobin in pulmonary capillary blood (cc/min/mmHg). The
infusion. Potential infusion complications included pulmonary vascu- MCID for improvement or worsening is a change in the % predicted
lature clotting and worsening cough or dyspnea caused by cell scale of 11 percentage points from baseline [52].
clumping or activation of the coagulation pathway. After discharge,
patients were followed for (S)AE by telephone weekly for 3 weeks, Imaging
and then at each in-person follow-up (1, 3, 6 and 12 months). (S)AEs CT scans of the lung were conducted pre-treatment, at 6 months
were categorized and graded per the Common Terminology Criteria and at 12 months using high-resolution CT imaging with a Phillips
for Adverse Events, Version 5.0. [47] In general, grade 1 AE comprise 128 scanner at 3-mm slices. CT images were acquired in the supine
mild symptoms; grade 2 is moderate symptoms; grade 3 is severe position at full inspiration. The CT findings representing sequelae of
symptoms; grade 4 is life-threatening symptoms; and grade 5 is COVID-19 pneumonia have been well-documented with parenchy-
death. mal changes described as ground-glass pattern, beehive pattern,
reticular pattern, traction bronchiectasis, septal thickening and archi-
Clinical utility monitoring and assessments tectural distortion [53,54]. Similar findings for pulmonary fibrosis
have also been documented [55,56].
Subjects were seen in-person follow-up visits at 1, 3, 6 and 12 The extent of lung fibrosis in each patient was then documented,
months. using a semiquantitative scoring system as described by Xie and
Zhao [57] in which the degree of involvement in individual lung seg-
Pulmonary function testing ment is assessed. Using fixed anatomic criteria, each lung was divided
Spirometry and DLCO were used to evaluate pulmonary function into three zones: upper zone (above the carina), middle (below the
mechanics (volumes and flows) and alveolar-capillary oxygen diffu- carina to the inferior pulmonary vein) and lower (below the inferior
sion, measured at baseline, 1, 3, 6 and 12 months. Both observed val- pulmonary vein). Scores were then assigned to each lung zone, as fol-
ues and % predicted scores were recorded. Published minimum lows: 0 = 0% involvement; 1 = less than 25% involvement; 2 = 25% to
clinically important difference (MCID) for improvement or worsening no less than 50% involvement; 3 = 50% to 75% involvement and
for pulmonary function tests (PFTs) were used to evaluate changes 4 = 75% or greater involvement. Each patient lung can have a maxi-
over time. mum score was 4 £ 3 = 12. All CT scans were read and scored by a
FVC is a measure of the maximum exhalation volume (liters) after single study team radiologist.
maximum inhalation. The MCID for improvement or worsening is a
change in the % predicted scale of 6 percentage points from baseline Results
[48].
Forced expiratory volume (FEV1) is a measure of the maximum Safety
volume in liters that can be expired in 1 second after maximum inspi-
ration. The MCID for improvement or worsening is a change in the % Subjects were followed in the hospital overnight with vital signs
predicted scale of 12 percentage points from baseline [49]. and clinical monitoring for potential pulmonary (S)AE related to the
TLC is a measure of the volume (liters) in the lungs at maximum SVF cell infusion (anaphylactic reaction; skin rash, cough, dyspnea,
inflation. Average values for TLC are 5700 cc for men and 4200 cc for wheezing, desaturation or pulmonary embolism). None were
1080 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083

observed. No significant changes in blood pressure, heart rate, respi- Each lung zone could have a maximum score of 4 (75% [or more]
ratory rate, oxygen saturation or pulmonary symptoms were docu- of parenchyma with fibrotic changes), so each lung (right/left)
mented in any of the subjects. can have a maximum score of 12.
After discharge, patients were contacted by telephone weekly for CT abnormalities improved in subjects 1, 3, 4 and 5. The CT abnor-
3 weeks to assess S(AE). No pulmonary (S)AE events were reported in malities for subject 2 with a history of pneumonia at 3 months post-
the weekly follow-up period. Specifically, no worsening was noted in SVF infusion did not markedly improve at 12 months. A representa-
pre-existent symptoms of cough or dyspnea. The lipoaspiration tissue tive CT images—baseline and at 12 months—is shown for subject 3
harvest procedure was well tolerated by all patients. No infections, (Figure 2). To the degree that the CT abnormalities represent paren-
hematomas or seroma AEs were reported. Grade 1 (mild) liposuc- chymal fibrosis, reductions in the fibrotic patterns could correlate
tion-related AEs—postoperative pain, bruising and cutaneous sensi- with improvements in lung volumes (FVE1, FVC and TLC) at the
tivity at the lipoaspiration site—were present in all patients (as macro level, whereas on a micro level, reductions in perialveolar
expected) but resolved (as expected) in 3 weeks. scarring could correlate with improved alveolar capillary oxygen
Subjects were evaluated for (S)AE in-person at 1-, 3-, 6- and diffusion gradient (DLCO).
12-month follow-up visits. No intervention-related (S)AEs were
reported. Subject two did develop a presumed bacterial pneumo- Discussion
nia (grade 3) at 3 months with worsened cough and dyspnea. She
was treated at home with intravenous antibiotics with clinical This is the first clinical report, to our knowledge and per PubMed,
improvement within 1 week. The pneumonia and accompanying of using SVF to treat the pulmonary symptoms associated with long
worsened cough and dyspnea were considered by her physician COVID-19. In this study of five patients with long COVID-19 pulmo-
to be unrelated to the SVF treatment. No grade 4 or grade 5 AEs nary symptoms, 40M autologous SVF cells were administered intra-
were reported over the 12-month follow-up period. Subject five venously on a “same-day” harvest-isolate-administer protocol [25].
was lost to follow-up after 6 months when he left Nicaragua. AEs The primary outcome of the study was to evaluate the safety of the
are described in Table 2. SVF infusion. There were no significant intervention-related AEs dur-
ing the 12-month study. Subject 2 did have a presumed bacterial
Assessments for potential efficacy pneumonia (with worsened dyspnea and cough symptoms) at 3
months’ post-SVF infusion, considered unlikely per the subject’s phy-
Pulmonary function tests sician to be related to the SVF infusion.
This is a small study (N = 5) of a single treatment group with longi- The secondary study objective was to evaluate for potential evi-
tudinal pulmonary function data from baseline to 12 months, except dence of SVF efficacy on the basis of changes in the subject’s pulmo-
for subject 5, who was lost to follow-up after 6 months. At baseline, nary symptoms, PFTs and CT images. The subjects were, on average,
all 5 subjects presented with pulmonary symptoms that affected 163 days (range 76 257 days) from hospital admission to SVF treat-
their ability to manage activities of daily living, ambulation, and ment. Pulmonary symptoms improved in 1 month for cough in three
work activities. Tables 3-6 provide the per subject observed and pre- subjects, whereas dyspnea improved in 3 months for three subjects
dicted values for all four PFT measurements from baseline to 12 after SVF infusion. Four subjects had improvements in % FVC pre-
months. dicted at 1-month post-SVF infusion that persisted or further
improved out to 12 months, whereas three subjects (subjects 1, 3,
Imaging and 4) had improvements in FEV1 % predicted at 1 or 3 months’ post-
Anatomic abnormalities in the pulmonary parenchyma were SVF infusion and maintained the improvement through the study
readily detected by CT scans. The predominant CT findings at period. The same three patients had improvements in % predicted
baseline (pre-SVF treatment) were the reticular and ground-glass DLCO at 1 month post-SVF infusion and continued to improve out to
patterns and traction bronchiectasis, all a (presumed) result of 12 months. Furthermore, these findings were re-enforced by
COVID-19 induced fibrosis. As described in Methods section, improvements in their CT lung parenchyma scores at 12 months for
each subject’s lung was divided into three zones and parenchyma four subjects. Bellan et al. [12,13] did report that 20% of patients with
pattern scores were then assigned to each lung zone (Table 7). an altered DLCO at 4 months’ post-hospital discharge showed a

Table 2
Adverse events.

Event grade (CTCAE v5.0)

Symptom (event) Pre-SVF baseline 1 week 1 month 3 months 6 months 12 months Procedures or SVF related

Subject Dyspnea
1 1 1 1 0 0 0
2 2 2 1 2 2 2 Unlikely
3 1 1 1 1 0 0
4 2 2 1 0 0 0
5 1 1 1 0 0 NA
Subject Cough
1 1 1 0 0 0 0
2 2 2 1 2 2 2 Unlikely
3 2 2 1 1 0 0
4 1 1 1 0 0 0
5 1 1 0 0 0 NA
Subject Lung infection
2 0 0 0 3 0 0 Unlikely
Subject Post-lipoaspirate bruising, pain
All 0 1 0 0 0 0 Yes
CTCAE, Common Terminology Criteria for Adverse Events; NA, not available; SVF, stromal vascular fraction.
 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083 1081

Table 3
FVC observed and % predicted values.

Subject FVC observed, L FVC % predicted

Mo 0 Mo 1 Mo 3 Mo 6 Mo 12 Mo 0 Mo 1 Mo 3 Mo 6 Mo 12

1 2.07 2.30 2.33 2.50 2.55 64 71 76 82 82

2 1.67 1.69 1.68 1.68 1.59 67 68 68 68 63
3 1.97 2.48 3.09 3.06 3.41 46 58 73 73 81
4 2.03 2.94 3.07 3.25 3.40 43 62 65 69 72
5 2.92 3.25 3.20 3.51 NA 58 64 63 67 NA
Given an MCID improvement for FVC % predicted of 6%, four subjects were improved by 1 month and
maintained an improved FVC for 6 12 months. Subject 2 (with a presumed bacterial pneumonia at
3 months) demonstrated no change across the 12 months.
FVC, forced vital capacity; MCID, minimal clinically important difference; NA, not available.

Table 4
FEV1 observed and % predicted values.

Subject FEV1 observed FEV1 % predicted

Mo 0 Mo 1 Mo 3 Mo 6 Mo 12 Mo 0 Mo 1 Mo 3 Mo 6 Mo 12

1 1.92 2.11 2.17 2.14 2.24 72 80 88 86 89

2 1.50 1.44 1.50 1.50 1.41 78 75 77 77 71
3 1.74 2.25 2.82 2.78 3.96 51 56 83 83 91
4 1.76 2.49 2.61 2.76 2.91 46 65 68 72 76
5 2.56 2.69 2.80 3.04 NA 61 65 67 71 NA
Given an MCID improvement for FEV1 12% for % predicted, 3 subjects were improved at 12 months, with
subject 5 trending toward improvement at 6 months. Subject 2 demonstrated no change across the
12 months.
FEV1, forced expiratory volume in 1 second; MCID, minimal clinically important difference; NA, not avail-
able.

Table 5
TLC observed and % predicted values.

Subject TLC observed, L TLC % predicted

Mo 0 Mo 1 Mo 3 Mo 6 Mo 12 Mo 0 Mo 1 Mo 3 Mo 6 Mo 12

1 3.49 3.75 3.94 3.59 4.25 78 84 95 87 101

2 2.62 3.21 2.96 2.96 2.90 76 94 86 86 81
3 2.81 3.86 4.61 4.87 5.40 52 71 85 90 99
4 4.09 5.03 4.97 5.39 5.37 66 81 80 87 86
5 3.98 5.09 4.97 5.46 NA 61 76 72 80 NA
An MCID for TLC is not clearly defined. All subjects had severe COVID-19, and all had a TLC predicted <80%.
The four subjects without a subsequent pneumonia all showed an increase to the TLC. Subject 2 initially
had increase at month 1 but worsened with the pneumonia.
COVID-19, coronavirus disease 2019; MCID, minimal clinically important difference; NA, not available;
TLC, total lung capacity.

Table 6
DLCO observed and % predicted values.

Subject DLCO observed, cc/min/mm Hg DLCO % predicted

Mo 0 Mo 1 Mo 3 Mo 6 Mo 12 Mo 0 Mo 1 Mo 3 Mo 6 Mo 12

1 10.6 13.4 11.8 13.2 16.3 46 59 54 60 74

2 09.5 06.5 07.3 07.3 08.2 51 34 39 39 43
3 10.5 15.8 18.8 25.0 29.6 34 51 61 82 97
4 10.7 22.7 27.1 30.0 31.2 32 68 81 90 94
5 19.3 23.3 19.9 23.0 NA 54 66 56 64 NA
Given an MCID improvement for DLCO percent-predicted change of 11%, at 12 months, three subjects
were improved, whereas subject 5 was improved at 6 months. Subject 2’s DLCO % predicted was worse at
12 months but did not meet the MCID worsened percent-predicted change of 11%.
DLCO, diffusing capacity of the lungs for carbon monoxide; MCID, minimal clinically important difference;
NA, not available.

significant improvement in DLCO at 1 year whereas 29% showed a PFTs improved for four subjects within 1 to 3 months after SVF
significant worsening. With only four subjects, and with other treatment is encouraging.
studies indicating clinical improvement in pulmonary symptoms Although drawing conclusions from this study is limited by its
without specific interventions, these results can only be consid- small size, the study’s results (safety and potential efficacy) are simi-
ered as hypothesis-generating for the role of SVF in improving lar to results with cultured AD MSC recently published. Vij et al. [58]
long COVID symptoms. That said, the fact that symptoms and evaluated 10 subjects with post-COVID-19 syndrome. Ten subjects
1082 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083

Table 7
Individual lung parenchyma scores.

Time, month Subject 1 2 3 4 5

Right Left Right Left Right Left Right Left Right Left

T=0 Score per level 4 4 1 2 3 4 4 3 3 4

4 4 1 2 3 4 2 4 3 4
4 4 1 1 3 4 2 3 2 4
Total 12 12 3 5 9 12 8 10 8 12
T=6 Score per level 4 4 1 1 3 2 1 2 1 1
4 4 1 1 2 2 1 2 1 1
1 1 1 1 2 2 1 1 0 0
Total 9 9 3 3 7 6 3 5 2 2
T = 12 Score per level 4 4 1 1 0 0 0 0
4 4 1 1 1 1 0 0
1 1 0 0 0 0 1 1
Total 9 9 2 2 1 1 1 1

Fig. 2. At the level of the right and left primary bronchi, reticular pattern and ground-glass patterns are seen, mild-to-moderate bronchiectasis is present in both lung parenchyma,
with a predominance on the right side. At 12 months, these specific findings were mostly resolved.

with post-COVID-19 syndrome were treated with multiple doses Funding

(200 £ 106) of autologous, commercially produced, AD MSC. Mild AEs
were noted but no (S)AEs were observed. Improvements were evi- The authors thank the Ministry of Health of Nicaragua and the
dent in quality of life and fatigue assessments. Further clinical trials Hospital Escuela Oscar Danilo Rosales Argu€ ello at the National Auton-
for safety and efficacy, using “same-day” autologous SVF adminis- omous University of Nicaragua for their support.
tered by intravenous infusions to treat long COVID with pulmonary
symptoms (and other symptoms), should be conducted. Author Contributions

Conception and design: MHC, KJB. Acquisition of data: MHC, JT,

YD, CR, JL, CL. Analysis and interpretation: MHC, KJB. Drafting and
Conclusions
revising: MHC, KJB.
This case series documents the safety of a single treatment with
autologous AD SVF cells via intravenous infusion for patients with Acknowledgments
long-term pulmonary sequelae of COVID pneumonia. No significant
short- or long-term AEs were associated with this intervention. Three The authors acknowledge the additional contributions of Dr. Sam-
measured parameters of pulmonary function (FVC, FEV1 and DLCO) uel Vilchez. Donation of SVF-2 devices for this study by GID Bio also
improved more than MCID for four of the subjects. These responses is appreciated.
suggest a potential therapeutic effect that was durable over the 12-
month follow-up period. Further investigations with SVF cellular References
therapy for post COVID-19 pulmonary sequelae (and other forms of
[1] World Health Organization. WHO coronavirus (COVID-19) dashboard, <https://
pulmonary fibrosis) are warranted to further evaluate efficacy, per- covid19.who.int> [accessed 31.01.2024].
form dose escalation studies, and further elucidate potential mecha- [2] World Health Organization. Post COVID-19 condition (long COVID),
nisms of action. <https://www.who.int/europe/news-room/fact-sheets/item/post-covid-19-
condition.> [accessed 25.09.2023].
TagedAP RListIem TagedAP RListabel[3]TagedAPTREn Harvard Scholar. The economic cost of long COVID: an update, <https://scholar.
harvard.edu/files/cutler/files/long_covid_update_7-22.pdf> [accessed 25.09.
2023].
Declaration of Competing Interest [4] Yong SJ. Long COVID or post-COVID-19 syndrome: putative pathophysiology, risk
factors, and treatments. Infectious Dis (Lond) 2021;53(10):737–54.
MC is a consultant for GID Bio. All other authors have no commer- [5] Ackerman M, Verleden SE, Kuehnel M, et al. Pulmonary vascular endothelialitis,
thrombosis and angiogenesis in Covid-19. N Eng J Med 2020;383:120–8.
cial, proprietary, or financial interest in the products or companies [6] Barton LM, Duval EJ, Stroberg, et al. COVID autopsies, Oklahoma, USA. Am J Clin
described in this article . Pathol 2020;153:725–33.
 M. Carstens et al. / Cytotherapy 26 (2024) 1076 1083 1083

[7] Menter T, Haslbauer JD, Reinhold R, et al. Post-mortem examination of COVID-19 [34] de Dios C, Vij R, Kim H, Park H, Chang D. Safety of multiple intravenous infusions
patients reveals diffuse alveolar damage with severe capillary congestion and of adipose-derived mesenchymal stem cells for hospitalized cases of COVID- 19: a
variegated findings of lungs and other organs suggesting vascular dysfunction. randomized controlled trial. Front Med. 2023;10:1321303.
Histopathology 2020;77:198–209. [35] Gentile P, Calabrese C, DeAngelis B. Impact of different preparation methods to
[8] Wichmann D, Sperhake J-P, Lu € tgetehtmann M, et al. Autopsy findings and venous obtain human adipose-derived stromal vascular fraction cells (AD-SVFs) and
thromboembolism in patients with COVID-19: A prospective cohort study. Ann human adipose-derived mesenchymal stem cells (AD-MSCs): enzymatic diges-
Int Med 2020;173:268–77. tion versus mechanical centrifugation. Int J Mol Sci 2019;20:21.
[9] Mo X, Jian W, Su Z, et al. Abnormal pulmonary function in Covid-019 patients at [36] Gentile P, Sterodimas A, Pizzicannella J, et al. Research progress on mesenchymal
time of hospital discharge. Eur Resp J 2020;55:2001271. stem cells (MSCs), adipose-derived mesenchymal stem cells (AD-MSCs), drugs,
[10] Thomas M, Price OJ, Hull JH. Pulmonary function and Covid-19. Curr Opin Physiol and vaccines in inhibiting COVID-19 disease. Aging Dis 2020;1(5):1191–201.
2021;21:29–36. [37] Gentile P, Sterodimas A. Adipose-derived stromal cells (ASCs) as a new regenera-
[11] Foti G, Longo C, Facciolii N, et al. Quantitative assessment of lung volume and tive immediate therapy for combating coronavirus (COVID-19)-induced pneumo-
enhancements in patients with Covid: role of dual-energy CT. Diagnostics nia. Exp Opinion Biologic Ther 2020;20(7):711–6.
2023;13:1201–12. [38] Kapur SK, Dos-Anjos Vilaboa S, Llull R, Katz AJ. Adipose tissue and stem and pro-
[12] Bellan M, Soddu D, Balbo PE, et al. Respiratory and psychophysical sequelae genitor cells: discovery and development. Clin Plast Surg 2015;42:155–67.
among patients with COVID-19 four months after hospital discharge. JAMA Netw [39] Brown JC, Shang H, Li Y. Isolation of adipose-derived stromal vascular fraction
Open 2021;4(1):e2036142. cells using a novel point-of-care device: cell characterization and review of the
[13] Bellan M, Apostolo D, Albe  A, et al. Determinants of long COVID among adults literature. Tiss Eng Part C Methods 2017;23(3):125–35.
hospitalized for SARS-CoV-2 infection: a prospective cohort study. Front Immunol [40] Nguyen A, Guo J, Banyard DA, et al. Stromal vascular fraction: a regenerative real-
2022;13:1038227. ity? Part 1: concepts and review of the literature. J Plast Reconstr Aesthet Surg
[14] Friedenstein AJ, Petrakova KV, Kurolesova AI, Frolova GP. Heterotopic of bone 2016;69(2):170–9.
marrow. Analysis of precursor cells for osteogenic and hematopoietic tissues. [41] Guo J, Nguyen A, Widgerow AD. Stromal vascular fraction: a regenerative reality?
Transplantation 1968;6(2):230–47. Part 2: mechanisms of regenerative action. J Plast Reconstr Aesthet Surg
[15] Caplan AI. Mesenchymal stem cells. J Orthop Res 1991;9(5):641–50. 2016;69:180–8.
[16] Wang S, Lei B, Zhang E, et al. Target therapy for inflammatory diseases with mes- [42] Zhu M, Xue J, Lu S, et al. Anti-inflammatory effect of stromal vascular fraction cells
enchymal stem cells and their derived exosomes: from basic to clinics. Int J in fat transplantation. Exp Ther Med 2019;17:1435–9.
Nanomed 2022;17:1757–81. [43] Bowles AC, Wise RM, Gerstein BY, et al. Immunomodulatory effects of adipose
[17] Gregoire C, Laylis N, Lambermont B. Bone marrow-derived mesenchymal stromal stromal vascular fraction cells promote alternative interaction macrophages to
cell therapy in severe COVID-19: preliminary results of a phase I/II clinical trial. repair tissue damage. Stem Cells 2017;35:2198–207.
Front Immunol 2022;13:932360. [44] Carstens MH, Zelaya KM, Calerio D, et al. Adipose-derived stromal vascular frac-
[18] Behnke J, Kremer S, Shazad T, et al. MSC based therapies—new perspectives for tion (SVF) cells for the treatment of non-reconstructable peripheral vascular dis-
the injured lung. J Clin Med 2020;9:682–727. ease in patients with critical limb ischemia: a 6-year follow up showing durable
[19] Rogers CJ, Harman RJ, Bunnell BA, et al. Rationale for clinical use of adipose- effects. Stem Cell Res 2020;49:102071.
derived mesenchymal stem cells for COVID-19 patients. J Translational Med [45] Carstens MH, Quintana FJ, Calderwood ST, et al. Treatment of chronic diabetic foot
2020;18:203–22. ulcers with adipose-derived stromal vascular fraction injections: safety and evi-
[20] Aghayan HR, Saliman F, Abedini A, et al. Human placenta-derived mesenchymal dence of efficacy at 1 year. Stem Cells Transl Med 2021;10(8):1138–47.
stem cells transplanted in patients with acute respiratory distress syndrome [46] Garza J, Campbell RE, Tjournakaris FP, et al. Clinical efficacy of intra-articular mes-
(ARDS) caused by COVID-19 (phase 1 clinical trial): safety profile assessment. enchymal stromal cells for the treatment of knee osteoarthritis: A double-
Stem Cell Res Ther 2022;13(1):365–77. blinded, prospective, randomized, controlled clinical trial. Am J Sports Med
[21] Hashemin S-MR, Aliannejad R, Zarrabi M. Mesenchymal stem cells derived from 2020;489(3):588–98.
perinatal tissues for treatment of critically-ill ARDS patients: a case series. Stem [47] National Cancer Institute. Common Terminology Criteria for Adverse Events
Cell Res Ther 2021;12:91–103. (CTCAE), <https://ctep.cancer.gov/protocolDevelopment/electronic_applications/
[22] Zheng G, Huang L, Tong H, et al. Treatment of acute respiratory distress syndrome ctc.htm#ctc_50> [accessed 25.09.2023].
with allogenic adipose-derived mesenchymal stem cells: a randomized, placebo- [48] DuBois RM, Wicker D, Albera C, et al. Forced vital capacity in patients with idio-
controlled pilot study. Respiratory Res 2014;15:39–48. pathic pulmonary fibrosis: test properties and minimal clinically important dif-
[23] Yao W, Shi Lei, Zhang Y, et al. Mesenchymal stem/stromal cell therapy for COVID- ferences. Am J Respir Crit Care Med 2011;184(12):1382–9.
19 pneumonia: potential mechanisms, currently clinical evidence, and future per- [49] Donohue JF. Minimal clinically important differences in COPD lung function.
spectives. Stem Cell Res Ther 2022;13:124–37. COPD 2005;2(1):111–24.
[24] Rebelatto CLK, Senegaglia AC, Franck CL, et al. Safety and long-term improvement [50] Noth I. Pulmonary function testing for pathologists. In: Leslie KO, Wick MR, eds.
of mesenchymal stromal cell infusion in critical COVID-19 patients: a randomized Practical pulmonary pathology: a diagnostic approach, 3rd ed., Elsevier;
clinical trial. Stem Cell Res Ther 2022;13:122. 2018:15–20.
[25] Shi L, Wang L, Xu R, et al. Mesenchymal stem cell therapy for severe COVID-19. [51] Lessa T, Pereira CA, Soares MR, et al. Reference values for pulmonary volumes by
Signal Transduct Target Ther 2021;6:339–44. plethysmography in a Brazilian sample of white adults. J Bras Pneumol 2019;45
[26] Shi L, YuaN X, Yao W, et al. Human mesenchymal stem cells treatment for severe (3):e20180065.
COVID-19: 1-year follow-up results of a randomized, double-blind, placebo-con- [52] Horita N, Miyazawa N, Kojima R, et al. Minimum clinically important differ-
trolled trial. EBioMed 2022;75:1037897. ence in diffusing capacity of the lungs for carbon monoxide in patients with
[27] Zuk PA, Zhu M, Ashjian P, et al. Human adipose tissue is a source of multipotent severe and very severe chronic obstructive pulmonary disease. COPD 2015;
stem cells. Mol Biol Cell 2002;13(12):4279–95. 12(1):31–7.
[28] Caplan AI, Correa D. The MSC: an injury drugstore. Cell Stem Cell 2011;9 [53] Kanne JP, Little BP, Schulte JJ, et al. Long-term lung abnormalities associated with
(1):11–5. COVID-19 pneumonia. Radiology 2023;306(2):e221806.
[29] Dicker A, Le Blanc K, Astro € m G, et al. Functional studies of mesenchymal stem [54] Wang Y, Dong C, Hu Y, Li C, Ren Q, Zhang X, Shi H, Zhou M. Temporal changes in
cells derived from adult adipose tissue. Exp Cell Res 2005;309:283–90. CT findings in 90 patients with COVID pneumonia: a longitudinal study. Radiol-
[30] Bourin P, Bunnell BA, Casteilla L, et al. Stromal cells from the adipose tissue- ogy 2020(March). https://doi.org/10.1148/radio.2020200843.
derived stromal/stem cells: a joint statement of the International Federation for [55] Ajlan AM, Ahyad RA, Jamjoom LG, Alharthy A, Madani TA. Middle East respiratory
Adipose Therapeutics and Science (IFATS) and the International Society for Cellu- syndrome coronavirus (MERS-CoV) infection: chest CT findings. AJR Am J Roent-
lar Therapy (ISCT). Cytotherapy 2013;15:641–8. genol 2014;203(4):782–7.
[31] Leng Z, Zhu R, Hou W, et al. Transplantation of ACE2- mesenchymal stem cells [56] Martinez-Sanchez 35 National Congress Spanish Society of Radiology Proceedings
improves the outcome of patients with COVID-19 pneumonia. Aging Dis 2020;11 2021; 1(1) <https://piper.espacio-seram.com/index.php/seram/issue/view/4>
(2):216–28. [accessed 1.06.2023].
[32] Shetty AK, Shetty AK. Mesenchymal stem cell infusion shows promise for combat- [57] Xie X, Zhong Z, Zhao W, Zheng C, Wang F, Lin J. Chest Ct for typical coronavirus
ing coronavirus (COVID-19)- induced pneumonia. Aging Dis 2020;11:462–4. disease 2019 (COVID-19) pneumonia. Relationship to negative RT-PCR testing.
[33] Sanchez-Guijo F, Garcia Arranz M, Lopez-Parra M, et al. Adipose derived Radiology 2020;296:E41–5.
mesenchymal stromal cells for the treatment of patients with severe SARS-CoV- [58] Vij R, Kim H, Park H, Cheng T, Lofti D, Chang D. Adipose-derived autologous mes-
pneumonia requiring mechanical ventilation. A proof-of concept study. EClinical- enchymal stem cell therapy for patients with post-COVID-19 syndrome: an inter-
Medicine 2020;25:100454. mediate-size expanded access program. Stem Cell Res Ther 2023;14:287.