Clinical Trial Details
— Status: Recruiting
Administrative data
NCT number |
NCT05982912 |
Other study ID # |
RMB-22-0398 |
Secondary ID |
|
Status |
Recruiting |
Phase |
|
First received |
|
Last updated |
|
Start date |
March 3, 2022 |
Est. completion date |
December 2025 |
Study information
Verified date |
August 2023 |
Source |
Rambam Health Care Campus |
Contact |
Ivan GUR, MD |
Phone |
0542555655 |
Email |
ostyly[@]gmail.com |
Is FDA regulated |
No |
Health authority |
|
Study type |
Observational
|
Clinical Trial Summary
The diagnosis of invasive pulmonary aspergillosis (IPA) bears grave implications for the
prognosis and treatment plan of the immunosuppressed patient. Thus far, such diagnosis in the
immunosuppressed patient, such as patients with acute myeloid leukemia (AML), relied heavily
on chest computed tomography (CT) and bronchoalveolar lavage (BAL), an invasive approach
bearing many caveats. Volatile organic compounds (VOC) are compounds that could be detected
in exhaled air, and have shown some potential in the non-invasive diagnosis of various
conditions, including IPA.
In this prospective longitudinal study we aim to compare the VOC profiles of patients
diagnosed with AML (baseline) to the profile of the same patient diagnosed with IPA later on,
and to the post recovery profile in the same patient. This approach should resolve many of
the issues plaguing prior attempts at VOC based IPA diagnosis, mainly the lack of properly
designed controls.
Samples will be collected from consenting patients using Tedlar bags, and analyzed using
thermal desorption gas chromatography mass spectrometry (TD-GC-MS). VOCs detected will be
digitally analyzed to construct different classification models, with predictive performances
compared to the clinical diagnosis using the accepted methods will be assessed by binary
logistic regression.
Description:
BACKGROUND The diagnosis of invasive pulmonary aspergillosis (IPA) bears grave implications
for the prognosis and treatment plan of the immunosuppressed patient. The omnipresence of
Aspergillus spp in the environment, and by extension - the ubiquity of such isolates in the
respiratory system of healthy subjects, poses a great diagnostic challenge. Namely, the
differentiation of IPA from non-invasive colonization of the respiratory tract, may be
challenging. The vast increase in prevalence of patients who are immunosuppressed, including
malignancies and oncological treatments, hematopoietic stem cells and solid organ
transplantations, and immunomodulatory medications, accentuates this dilemma.(1)
Particularly, patients with acute myelogenous leukemia (AML) are at high risk of developing
IPA, with 5-15% of such patients developing IPA during the first 3 months after AML
diagnosis.(1-3) As up to 5% of patients with AML may have IPA at presentation, patients
admitted to Rambam Medical Center (RMC) with a new diagnosis of AML routinely undergo chest
computerized tomography (CT), which serves as screening for the presence of IPA at diagnosis
and a baseline for further reference should IPA be suspected later on.(3) Diagnosis of
probable or proven IPA in an AML patient requires suggestive imaging findings along with
microbiologic evidence. While in some cases the diagnosis can be made using biomarkers
(mainly the d-galactomannan antigen and or Aspergillus spp polymerase chain reaction) in
peripheral blood, the sensitivity of these assays is low and often this minimally invasive
approach does not yield sufficient diagnostic certainty. In such cases, bronchoscopy with
bronchoalveolar lavage is usually performed. This invasive procedure is particularly
hazardous in patients with suspected IPA, the vast majority of whom are very frail due to
their underlying illness.(4) These considerations have led the quest for additional,
non-invasive modalities in the diagnosis of IPA. (5) One such approach has been volatile
organic compounds (VOC) analysis in exhaled air samples. Characterized by low vapor pressure
at room temperature, VOC profile analysis of exhaled air is an emerging field of diagnostics,
with the potential of non-invasive screening and diagnosis of various diseases, including
neoplasms,(6) infections(7) and inflammatory processes.(8) Previous studies have identified
potential VOC markers of metabolically active Aspergillus spp,(9) with one small prospective
study in humans(10) showing encouraging diagnostic performance for VOC based IPA
identification. However, these attempts have been hampered by several pitfalls. Firstly, VOC
profiles of in vitro Aspergillus spp seem to have little relation with VOCs found in breath
analysis of colonized patients.(5) Secondly, exhaled VOCs tend to differ greatly between
individuals, making the absence of temporal controls (i.e., VOC profile changes in the same
individual before and after IPA development) a crucial caveat.(10) Thirdly, to our knowledge
no longitudinal studies have assessed the response of VOC profiles to IPA treatment.
In this prospective longitudinal study we aim to overcome these hurdles taking advantage of
two key characteristics of hematologic patients treated in RMC. First, patients with newly
diagnosed AML have a relatively high prevalence of IPA. Second, a chest CT is performed at
the diagnosis of AML in all patients. Thus, collecting breath samples from patients with
normal baseline CT will serve as negative controls, allowing for the comparison of VOC
profiles before and after the development of IPA in the same patient. Further sampling after
the patient has recovered from IPA will provide additional longitudinal control.
GOALS The identification of VOC profiles predictive of IPA in breath samples
METHODS
Study design:
This is a single center, non randomized, prospective cohort study. Breath samples from
consenting patients will be analyzed for VOC profiles at the time of initial CT screening for
IPA. In addition, samples will also be collected at the time of diagnosis of any possible,
probable or confirmed IPA (defined according to the EORTC criteria (4) as detailed below).
Setting and participants:
Patients newly diagnosed with AML initiating treatment in the RMC hematology department will
be approached. Inclusion period will be from October 1, 2022, until March 31, 2025.
Inclusion criteria:
New diagnosis of acute myeloid leukemia AND/OR planned hematopoietic stem cells
transplantation (HCT) Chest CT performed within 30 days from sampling 18 years of age or
older The ability to provide tidal breath samples totalling 10L directly into a Tedlar bag
Exclusion criteria:
Any condition impairing the patient's ability to provide informed consent
Sample collection:
Breath samples will be collected in one of our hematological department's 23 dedicated
hospitalization rooms - positively pressurized and climate controlled with E-pure
high-efficiency particulate air (HEPA) filters (ADS-Laminaire, Israel). Patients will be
required to abstain from teeth brushing, dental wash, eating, drinking or smoking in the 6
hours prior to sample collection. After being seated for 10 minutes (to avoid exercise
induced isoprene concentration increase (11) ), the patient will be asked to fill 10L of
tidal breath into a tedlar bag (SKC, South Korea). Simultaneously, an identical tedlar bag
shall be filled with ambient air.
Upon each sample collection, the following data will be recorded:
Patient's characteristics, including age, hemato-oncological diagnosis, time from initial
diagnosis, current treatment, time from HCT if applicable, background diagnoses Current
pharmacological treatment, including all antibiotics and antifungals prescribed the preceding
month.
Current diagnostic tests, including all cultures, serology assays, nucleic acid amplification
studies, complete blood count and chemistry, voriconazole plasma level and chest imaging
performed during the past month.
Sample handling and VOC extraction:
All tedlar bags will be stored at 4℃ following collection. Adsorption onto 2 parallel thermal
desorption tubes shall be accomplished via Gilian LFS-113 Low Flow Personal Air Sampling Pump
(SRA instruments, Milano, Italy), performed within 72 hours from sampling and stored under
ambient conditions. These samples will be labeled (coded) and transformed to the
Environmental Chemistry Laboratory (Kasali Institute of Chemistry, Hebrew University of
Jerusalem, Israel)
Analytics:
Thermal desorption will be accomplished at 290 for 20 minutes using analytical grade
(99.999%) helium gas as carrier. Injection to tandem gas chromatography mass spectroscopy
(GC-MS) shall be performed at similar inlet temperatures not exceeding the flow of 40 mL/min.
The initial GC temperature program will mimic previously reported (10) protocols: 40˚C for 3
minutes, raised to 70˚C at a rate of 5˚C per minute and held for 3 minutes, raised to 203˚C
at 7˚C per minute and held for 4 minutes, raised rapidly to 270˚C and held for 5 minutes. A
triple quad mass spectrometer (Agilent Technologies, Santa Clara, CA) shall be connected in
tandem, with a range of mass measurement of 20-400 m/z.
The diagnosis of Invasive Pulmonary Aspergillosis will be defined according to the updated
EORTC criteria (4) as either one of the following:
Confirmed IPA defined as a respiratory biopsy specimen with positive culture of Aspergillus
spp or histopathologic demonstration of tissue invasion by hyphae.
"Probable IPA", defined (1) as compatible radiographic features on chest CT (Dense,
well-circumscribed lesion(s) with or without a halo sign, air-crescent sign or a cavity) and
any of the following: Evidence of Aspergillus spp upon cytology, direct microscopy or culture
from any respiratory specimen, including sputum, broncho-alveolar lavage (BAL) fluid or
bronchial brush Positive D-galactomannan antigen detected in plasma, serum, or BAL fluid
Positive polymerase chain reaction (PCR) for Aspergillus spp detected in plasma, serum, or
BAL fluid "Possible IPA", defined (1) as compatible radiographic features on chest CT (Dense,
well-circumscribed lesion(s) with or without a halo sign, air-crescent sign or a cavity)
without corroboratory laboratory findings.
Additional information regarding the probability of IPA diagnosis will be collected by two
independent observers (I.G. and A.S.) reviewing the full electronic medical record (EMR) at
the end of the study period, assigning a single probability score 1-10 (1 being very low
probability of IPA, 10 corresponding to almost certain IPA) in view of the full clinica
course of the disease.
Additional variables and data sources: For all cohort patients, epidemiological and clinical
information will be extracted from the hospital's electronic medical records (EMR) system
used in our hospital retrieved with the assistance of the MD-clone software. We will need to
obtain the personal information of patients in order to access their data recorded in the
daily follow up in their medical chart as some data is not coded. All data shall be later
coded and stripped off any personal identifiers including name, surname, date of birth, ID
number, and address. All data shall be stored on a designated hard disk at the hands of the
primary investigator. Data sources will include the patient's file on the "prometheus" EMR
registry as well as retrieved data using MD-clone software. Collected data will include:
demographics (age, sex), background diagnoses, length of stay and transfer to other units,
laboratory results, vital signs recorded and the oxygenation method used as documented in the
daily follow-up.
Sample collection schedule: The first sample of consenting patients included in this study
will be collected within 7 days from baseline chest CT. Subsequently, following any clinical
suspicion of IPA by the treating physician - a suspicion that invariably leads to a chest CT
being performed, samples will be collected within 7 days from chest CT. Additional samples
will be collected within 7 days from the discontinuation of antifungal treatment by the
treating physician. The sampling schedule, are visualized in figure 1.
Figure 1 - Sampling strategy. This study's intervention is limited to sampling underlined in
bold red - all other steps are part of the usual management of AML patients and are in no way
affected by the participation in this study. AML - Acute Myeloid Leukemia; Dx - Diagnosis; CT
- Computerized Tomography; IPA - Invasive Pulmonary Aspergillosis; Tx - Treatment; d/c -
discontinued; GGO - Ground Glass Opacities.
Data handling and statistical analysis:
Clinical data shall be analyzed using descriptive statistics, parametric and nonparametric
tests as appropriate. All statistical analysis shall be performed using R software version
4.0.0 ("Arbor day") or later. All identifiable data, including patient's baseline
characteristics, laboratory and imaging results and previous diagnoses and treatment records,
shall be stored by the primary investigator on a designated RAS encrypted solid phase memory
device for the duration of the study. Patient's names, ID numbers, address and date of birth
will not be stored.
Mass spectroscopy outputs will be reviewed using the MassHunter workstation and software
(Agilent Technologies, Santa Clara, CA). Partial least squares discrimination analysis will
be preformed using the SIMCA software version 14.0 or later (MKS Umetrics, Sweden). For each
m/z range, a logistic regression model will be constructed, allowing for the calculation of
the variable importance for projection (VIP). For all VIPs ≥1 the corresponding ion profile
will be selected as a potential target. These will be compared with previously implicated
metabolites indicative of IPA, (10,11) as well as our existing database, as summarized in
APPENDIX A. Such identification is crucial since the same m/z value can correspond to
different metabolites, as different substances may have the same germplasm charge ratio.
Next, all identified substances will be confirmed against the National Institute of Standards
and Technology (NIST) 11 Mass Spectral Library (Scientific Instrument Services, Ringoes, New
Jersey) and a pure chemical standard. Finally, the digital matrix composed of the selected
characteristic VOCs will be imported into python (version 3.10.5 or later) to construct
different classification models, with predictive performances assessed by binary logistic
regression. The correlations between exhaled breath VOC expression levels and levels of serum
and BAL d-gallactomannan and IPA diagnostic certainty (confirmed, probable or possible, as
defined below) will be accomplished using Person's r.
Sample size and power:
In this prospective longitudinal study patients with high probability of developing IPA
(i.e., patients diagnosed with acute leukemia) will be enrolled at diagnosis. Since our
center performs baseline chest CT screening to identify potential IPA at baseline, VOC
collected after normal radiography can serve as ample controls. At least 15% of such patients
go on to develop IPA in the ensuing months,(3) limiting the sample size needed at
recruitment. Based on previously reported adherence to follow-up,(12) we expect drop out rate
to be low, allowing for the detection of previously elusive post treatment changes in VOC
profiles.
For the purpose of these calculations we assumed the previously reported VOC assays
sensitivity (compared to microbiology or antigen testing in respiratory secretions, BAL or
patient serum) of 80-95% each.(10) Recent retrospective surveys have identified the incidence
of IPA to be 5% at initial screening CT and additional 10% within the initial
hospitalization.(3) This translates into an estimated sample size of 124 patients, calculated
under the assumption of the accepted significance (alpha = 0.05) and power (beta = 0.8).
Allowing for an estimated 70% enrolment and the additional 15% exclusion rate, the total
estimated sample size is 208 patients. Projecting the average annual incidence of 90 new
cases of acute leukemia treated in our hospital, this sampling goal should be easily reached
within the prespecified 30 months recruitment period. We thus expect the number of
participants to be 300.
ETHICAL CONSIDERATIONS The development and amelioration of noninvasive techniques for IPA
identification has been the stated goal of various professional societies for over two
decades. (1,4) The patients proposed diagnostic tool bears minimal to no inconvenience to the
participants, and we deem potential side effects and danger to be negligible in this entirely
non-invasive modality.
Following data extraction, all data shall be stripped of any identification including name,
surname, ID number, case number, date of birth and address - all destroyed when initial data
extraction is completed. Data shall be stored securely on a RAS encrypted, designated SD
device at the hands of the PI, minimizing the probability of this study affecting data
privacy or medical confidentiality of the patients included. In view of the direct effect the
results of this inquiry may have on the management and the prognosis of the severely ill, we
find the benefits of this study to outweigh any potential risks.
Funding No external funding was provided for this study thus far.
Bibliography
1. Blot SI, Taccone FS, Van den Abeele A-M, Bulpa P, Meersseman W, Brusselaers N, et al. A
clinical algorithm to diagnose invasive pulmonary aspergillosis in critically ill
patients. Am J Respir Crit Care Med. 2012 Jul 1;186(1):56-64.
2. Souza L, Nouér SA, Morales H, Simões B, Solza C, Queiroz-Telles F, et al. Epidemiology
of invasive fungal disease in haematologic patients. Mycoses. 2021 Mar;64(3):252-6.
3. Bitterman R, Hardak E, Raines M, Stern A, Zuckerman T, Ofran Y, et al. Baseline Chest
Computed Tomography for Early Diagnosis of Invasive Pulmonary Aspergillosis in
Hemato-oncological Patients: A Prospective Cohort Study. Clin Infect Dis. 2019 Oct
30;69(10):1805-8.
4. Donnelly JP, Chen SC, Kauffman CA, Steinbach WJ, Baddley JW, Verweij PE, et al. Revision
and update of the consensus definitions of invasive fungal disease from the european
organization for research and treatment of cancer and the mycoses study group education
and research consortium. Clin Infect Dis. 2020 Sep 12;71(6):1367-76.
5. Savelieff MG, Pappalardo L. Novel cutting-edge metabolite-based diagnostic tools for
aspergillosis. PLoS Pathog. 2017 Sep 7;13(9):e1006486.
6. Tsou P-H, Lin Z-L, Pan Y-C, Yang H-C, Chang C-J, Liang S-K, et al. Exploring Volatile
Organic Compounds in Breath for High-Accuracy Prediction of Lung Cancer. Cancers
(Basel). 2021 Mar 21;13(6).
7. Kamal F, Kumar S, Edwards MR, Veselkov K, Belluomo I, Kebadze T, et al. Virus-induced
Volatile Organic Compounds Are Detectable in Exhaled Breath during Pulmonary Infection.
Am J Respir Crit Care Med. 2021 Nov 1;204(9):1075-85.
8. Castro-Rodriguez JA, Cifuentes L, Martinez FD. Predicting asthma using clinical indexes.
Front Pediatr. 2019 Jul 31;7:320.
9. Das S, Pal M. Review-Non-Invasive Monitoring of Human Health by Exhaled Breath Analysis:
A Comprehensive Review. J Electrochem Soc. 2020 Feb 3;167(3):037562.
10. Koo S, Thomas HR, Daniels SD, Lynch RC, Fortier SM, Shea MM, et al. A breath fungal
secondary metabolite signature to diagnose invasive aspergillosis. Clin Infect Dis. 2014
Dec 15;59(12):1733-40.
11. Li Z-T, Zeng P-Y, Chen Z-M, Guan W-J, Wang T, Lin Y, et al. Exhaled volatile organic
compounds for identifying patients with chronic pulmonary aspergillosis. Front Med
(Lausanne). 2021 Sep 23;8:720119.
12. Hardak E, Fuchs E, Leskes H, Geffen Y, Zuckerman T, Oren I. Diagnostic role of
polymerase chain reaction in bronchoalveolar lavage fluid for invasive pulmonary
aspergillosis in immunocompromised patients - A retrospective cohort study. Int J Infect
Dis. 2019 Jun;83:20-5.