The diagnostic and prognostic role of breath pattern ... · Santonico, Marco, et al. "Design and...

Post on 12-Aug-2020

0 views 0 download

Transcript of The diagnostic and prognostic role of breath pattern ... · Santonico, Marco, et al. "Design and...

The diagnostic and prognostic role of breath pattern analysis in the multimorbid

patient

Raffaele Antonelli Incalzi

Università Campus Bio Medico

Roma

Breath analysis: a topic of interest

Analysis of volatile organic compounds (VOCs): outline

• 1a. Methods: single gas measurement vs. assessment of breath pattern (BP)

• 1b. The e nose: an overview

• 2. BP in chronic non neoplastic diseases: Asthma, COPD, OSAS, Liver diseases, CHF

• 3. BP in neoplastic diseases

• 4. Impact of multimorbidity on BP

• 6. Short and long-term perspectives

The diagnostic and prognostic role of breath pattern analysis in the multi morbid patient

• 1a. Methods: single gas measurement vs. assessment of breath pattern (BP)

• 1b. The e nose: an overview

VOLATILE ORGANIC COMPOUNDS: FROM PRODUCTION TO EXHALED REMOVAL

Which VOCs are associated with a diagnosis of COPD? (Van Berkel J et al. Resp Med 2010; 104: 557)

VOCs associated with lung cancer

The E nose: how does it work?

GAS SENSOR ARRAY

• Quarzt Micro Balances functionalized with ANTHOCYANINS

- BIOLOGICAL SOURCE

- LOW COST

- TAILORING FOR SPECIFIC APPLICATIONS

- Natural pigments widely distributed in nature, belonging to the flavonoid group of polyphenols

- Anthocyanidins as such are highly unstable molecules

- The attachment of glycosyl units and acyl groups affect significantly both the stability and reactivity of the

anthocyanin molecule

Santonico, Marco, et al. "Design and Test of a Biosensor-Based Multisensorial System: A Proof of

Concept Study." Sensors 13.12 (2013): 16625-16640.

L. Buck and R. Axel; Cell 1991

S2 S3 S4 S5 S6 S7

n-pentane

propanaldehyde

methanol

ethanol

toluene

benzene

dimethylsulphide

tiophene

triethylamine

S1

acetic acid

Selectivity spectra of Natural & Artificial

Olfaction

The Bionote targets different gases at different temperatures

Breath sampling

European Patent 2641537 2013-09-25

• Pre-concentration of breath samples by absorption in adsorbent materials (Tenax), which reduces water vapour effect

European Patent 2641537 2013-09-25

EXPERIMENTAL SET-UP: GENERAL OVERVIEW

BREATHPRINT

Pennazza G. et al. Sensors and Actuators B: Chemical, 204, 1, 2014, Pages 578–587

The Bionote based BP: highly reproducible in the healthy subject (Antonelli Incalzi R et al. PLOS One 2012; 7: e45396)

The diagnostic and prognostic role of breath pattern analysis in the multimorbid patient

• 2.BP in chronic diseases: how multimorbidity affects the BP

• Evidence pertaining to:

• - COPD

• - Asthma

• - OSAS

• - Liver diseases

• - CHF

• - Neoplastic diseases

BP is fairly repeatible in a GOLD 4 COPD patient (Antonelli Incalzi R et al. PLOS One 2012; 7: e45396)

FEV1%=0.864

Might the BP substitute for spirometry in elderly unable to comply with quality standards? (Bellia V et al for the SaRA Study, Am J Respir Crit Care Med 2000; 161: 1094-1100)

BP predicts 6’ walked distance decline better than GOLD stage (Finamore P et al. Int J COPD 2018:13 1441–1448)

Asthma vs COPD (through Cyranose) (Fens N et al. Am J Respir Crit Care Med 2009; 180: 1076)

Asthma vs Controls and COPD vs smokers (Fens N et al. Am J Respir Crit Care Med 2009; 180: 1076)

The Cyranose: a valuable tool for the diagnosis of asthma… (Fens N et al. Am J Respir Crit Care Med 2009; 180: 1076)

… and to predict steroid responsiveness in steroid naive asthma patients (van der Schee et al.Clin. Exp. Allergy. 2013; 43: 1217–1225)

…but Cyranose derived BP is a weak correlate of asthma severity (Dragonieri S et al. J Allergy Clin Immunol 2007;120:856)

Clustering approach and effects of inhaled therapy on breath-

fingerprint of newly diagnosed COPD patients: an e-nose based

study (Scarlata S et al J Breath Res. 2018 Jun 8;12(3):036022 )

0

10

20

30

40

50

60 Q1

Q3 Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6

Q7 Q8

Q1 Q3

Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6

Q7 Q8

Radarplot with the breath-fingerprint of the three clusters before inhaled therapy

Cluster 1 Cluster 2 Cluster 3

The second cluster differs from others in shape, while the first and the third, albeit having the same shape, have differences in the area.

Cluster 3: the highest BODE index and greatest

Multimorbidity

Clustering approach and effects of inhaled therapy on breath-

fingerprint of newly diagnosed COPD patients: an e-nose based

study (Scarlata S et al J Breath Res. 2018 Jun 8;12(3):036022 )

0

10

20

30

40

50 Q1

Q3 Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6 Q7

Q8 Q1

Q3 Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6 Q7

Q8

LABA o LAMA o LABA/LAMA effects on breath-fingerprint

Naive Terapia

0

10

20

30

40 Q1

Q3 Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6 Q7

Q8 Q1

Q3 Q4

Q5

Q6

Q7

Q8

Q1

Q3

Q4

Q5

Q6 Q7

Q8

LABA/ICS effects on breath-fingerprint

Naive Terapia

Different effects of bronchodilators and corticosteroids on breath-fingerprint: LABA, LAMA or their combination alone did not affect the shape of the BP, but only were associated with a mild decrease in the BP area; on the other hand, inhaled steroids significantly modified the shape of the radar plot profile.

Effects of 3 month C-PAP ventilation on BP of OSAS (Greulich T et al. Eur Respir J. 2013;42:145)

One night ventilation changes the BP of OSAS, but in two alternative ways … (Antonelli Incalzi R et al. Sleep Breath 2015; 19: 623-30)

…depending upon the pattern of comorbidity of OSAS…

C D p

Diabetes mellitus n° (%) 5 (17.2) 9 (28) p<0.06*

Metabolic Syndrome n° (%) 8 (27.6) 14 (66.6) p<0.01*

Chronic Heart Failure n°

(%)

2 (6.9) 6 (28.6) p<0.05*

Atrial Fibrillation n° (%) 2 (6.9) 6 (28.6) p<0.05*

Number of comorbidities

(mean/standard deviation)

1.55 (1.0) 3.14 (1.8) p<0.01**

A normal BP is truly normal (Scarlata S et al. Sci Rep. 2017; 7: 11938)

The e nose for the screening of patients suitable for polysomnigraphy?

The e nose to select patients suitable for polysomnigraphy? Likely more for the elderly (Endeshaus Y. JAGS 2004; 52:957; JAGS 2006; 54: 1740-4)

E-NOSE E-NOSE +

CLINICA DATA

CLINICAL DATA

94% 84%

76%

E nose reliably identifies children exposed to parental smoke (Fasola S et al ATS 2018)

E nose:

BIONOTE

Breath collector:

Pneumopipe

Clinical data

Spirometry

The e nose in chronic liver diseases (De Vincentis A et al…Sci Rep 2016; May 5;6:25337)

(De Vincentis A et al…Sci Rep 2016; May 5;6:25337)

BP predicts survival and time to hospitalization in liver diseases (De Vincentis A et al. Liver International 2017; 37: 242-250. )

Radar plots of clusters (De Vincentis A et al. Liver International 2017; 37: 242-250. )

CHF vs controls and COPD (Finamore P et al Breath Res 2017, in press)

Reference

Prediction CHF* HC**

CHF 24 7

Controls 6 32

Reference

Prediction CHF* COPD***

CHF 19 9

COPD 11 26

*CHF: Congestive heart failure

**HC: Healthy controls ***COPD: Chronic obstructive pulmonary disease

Changes in BP as a function of improved (blue) vs unchanged/worsened NYHA class (red) (Finamore P et al Breath Res 2017, in press)

The diagnostic and prognostic role of breath pattern analysis in the multi morbid patient

• 5. The e nose in neoplastic diseases

N d r 1: only 5% of mutations leading to cancer are inherited mutations; the vast majority of them occurs during normal DNA replication. Furthermore, multiple and not isolated mutations are eventually responsible for cancerogenesis (Tomasetti C et al. Sci. Transl. Med. (2017); 355: 1330–1334). N d r 2: Liquid biopsy is a promising diagnostic tool, but low availability of circulating cells and ctDNA makes it frequently inconclusive in early stage cancer (Bettegowda et al. Sci Transl Med. 2014; 6: 224ra24)

Detection of lung cancer with volatile markers in the breath

Before and after melanoma resection (D’Amico A et al. Skin Research and Technology 2007..)

The Bionote for the screening of lung cancer (Rocco R et al. Eur J Cardiothorac

Surg 2015….)

High sensitivity for lung cancer detection using analysis of exhaled carbonyl compounds (Schumer EM et al. (J Thorac Cardiovasc Surg 2015;-:1-8)

• Methods: Patients with computed tomography–detected intrathoracic lesions and healthy control participants were enrolled from 2011 onward. One liter of breath was collected from a single exhalation from each participant. Concentrations of 2-butanone, 3-hydroxy-2-butanone,

• 2-hydroxyacetaldehyde, and 4-hydroxyhexanal were measured.

• Results: In all, 156 subjects had lung cancer, 65 had benign disease, and 194 were healthy controls. A total of 103 (66.0%) lung cancer patients were early stage (stage 0, I, and II). For 1 elevated cancer marker, breath analysis showed a sensitivity of 93.6%, and a specificity of 85.6% for lung cancer patients. Additionally, 83.7% of stage I tumors 2 cm were detected; whereas only 14% of the control population tested positive. In a comparison of cancer to benign disease, specificity was proportional to the number of elevated cancer markers present.

• Conclusions: Screening using a low-dose CT scan is associated with high cost, repeated radiation exposure, and low accrual. The high sensitivity, convenience, and low cost of breath analysis for carbonyl cancer markers suggests that it has the potential to become a primary screening modality for lung cancer.

The e nose (array based sensors) in the context of breath based detection of lung cancer (Queralto N et al. J. Breath Res. 8 (2014) 02711)

Evaluating Molecular Biomarkers for the Early Detection of Lung Cancer: When Is a Biomarker Ready for Clinical Use? An Official American Thoracic Society Policy Statement (Mazzone PJ et al. AJRCCM

2017; 196: e15-e29)

BP analysis: a strategy to limit negative effects of lung cancer screening? (Mazzone PJ et al. AJRCCM 2017; 196: e15-e29)

• Currently available and potential clinical use

The diagnostic and prognostic role of breath pattern analysis in the multi morbid patient

Potential clinical uses of the e nose

• Non neoplastic respiratory diseases • COPD: as a proxy of spirometry • COPD: as an index of obstruction

severity • Asthma: as a diagnostic tool, but

not to grade severity • COPD and Asthma: to monitor the

adherence to the therapy • OSAS: as a measure of efficacy of

and adherence to CPAP • OSAS: likely to select patients

amenable to PSG

• Non neoplastic and non respiratory diseases

• Liver diseases: to distinguish cirrhotic from non cirrhotic cases

• Liver diseases: as a primary prognostic index

• CHF: likely weak as a diagnostic tool

• CHF: possibly a marker of response to the therapy

• Other chronic diseases: inconclusive or very preliminary evidence

Potential clinical uses of the e nose

• Neoplastic diseases

• Lung cancer screening

• Monitoring the response to the therapy of selected cancers

• Early detection of recurrence of selected cancers

E nose: main limitations

• Breath collection needs to be standardized and optimized

• Different types of e nose should be compared and consistency of results checked

• The graphical output should benefit from being translated in a numerical and easy to understand form

• Effects of age on the BP are unknown

• Clinical trials suffer from the usually small sample size

• In most studies the BP of the target disease was studied only in the absence of confounding by multimorbidity

E nose: strengths

• Non invasive and no risk procedure

• Low cost (<10 euros)

• Suitable for point of care analysis

• Flexible due to the temperature dependent selective VOCs extraction

• Highly reproducible

• Very sensitive to metabolic changes

E nose: perspectives 1

• Close cooperation between bio-engineers and physicians to manage the purely technical issues.

• Definition of age-standardized BP

• To identify disease-specific VOCs in order to capitalize on the e nose flexibility

• Moving from mono/oligocentric to multicentric confirmatory studies

• Pointing at discriminating the metabolic from the inflammatory markers in BP

• In chronic diseases prone to acute exacerbations, to distinguish baseline or chronicity-related signals from overimposed ones.

E nose: perspectives 2

Acknowledgements A special thank to the engineers of the Campus Bio-Medico for their extraordinary scientific support

Unit of Electronics for Sensor Systems, Department of Engineering • Prof. Giorgio Pennazza, PhD

• Prof. Marco Santonico, PhD

• Simone Grasso, PhD

• And their teacher, Prof A D’Amico