Introduction

Inhaled hyperpolarized MRI provides a way to measure airway, alveolar and pulmonary vascular function in the lung simultaneously and with relatively high spatial resolution. In people with asthma, ventilation abnormalities are quantified as ventilation defect percent (VDP)¹ which has been shown to relate to airway inflammation^2,3 and remodeling,⁴ independently predict asthma control,⁵ and sensitively respond to therapy.^6,7 Recently, hyperpolarized ¹²⁹Xe MRI and spectroscopy (MRS) have made it possible to interrogate gas-exchange abnormalities via the detection of ¹²⁹Xe gas transfer into the alveolar-membrane and red-blood-cells (RBC).⁸ Abnormal gas-exchange has been quantified using the ratio of ¹²⁹Xe signal in the RBC and alveolar membrane (RBC:membrane).^9,10
Asthma is considered an airways disease, although, the presence of a vascular component has been shown using CT measurements of the pulmonary vascular tree.¹¹ Such pulmonary vascular abnormalities also normalized in patients with severe asthma following biologic therapy.¹² Here we aimed to investigate patients with moderate asthma and whether ¹²⁹Xe MRS gas-exchange measurements were related to 6-week response to ICS/LAMA/LABA (inhaled corticosteroid, long-acting muscarinic antagonist, long-acting beta-agonist).

Methods

Participants and Study Design:
Participants with moderate asthma provided written informed consent to an approved protocol to receive daily inhaled fluticasone furoate, umeclidinium, and vilanterol (200/62.5/25µg) for 6-weeks. Participants completed pre-/post-bronchodilator (BD) spirometry, oscillometry, fraction of exhaled nitric oxide (FeNO) and ¹²⁹Xe MRI/MRS at Week-0 and at Week-6.
MRI Acquisition:
Anatomical ¹H and hyperpolarized ¹²⁹Xe MRI/MRS were acquired using a 3T Discovery MR750 with broadband capability, as previously described.^13-15 Anatomical ¹H images were acquired using a fast-spoiled gradient-recalled echo (FGRE) sequence (total acquisition time=8s; TR msec/TE msec=4.7/1.2; flip-angle=30°; FOV=40×40cm²; bandwidth=24.4kHz; 128×80 matrix, zero-padded to 128×128; partial echo percentage=62.5%; 15-17 slices; slice thickness=15mm; no gap). Static ventilation images were acquired using a three-dimensional FGRE sequence (total acquisition time=14s; TR msec/TE msec=6.7/1.5; variable flip-angle; FOV=40×40cm²; bandwidth=15.63kHz; 128×128 matrix, zero-padded to 128×128; 14 slices; slice thickness, 15mm; no gap). ¹²⁹Xe MRS was acquired using a free-induction-decay whole-lung spectroscopy sequence (200 dissolved-phase spectra, TR msec/TE msec=15/0.7; flip-angle=40°; BW=31.25kHz, 600μs 3-lobe Shinnar-Le Roux pulse). Participants were instructed to inhale and hold 1.0L of gas (100% N₂ for anatomical scan, 400mL hyperpolarized ¹²⁹Xe + 600mL ⁴He for static ventilation scan, 200mL hyperpolarized ¹²⁹Xe + 800mL ⁴He for MRS scan) to ensure volume matched images. ¹²⁹Xe gas was polarized to 20%-50%.¹⁶
Data Analysis:
MRS peaks were fit to three complex Lorentzian distributions to determine compartment frequency, full-width half maximum, and area under the curve (AUC). RBC:membrane ratio was calculated as the ratio of RBC AUC to membrane AUC. Gas-transfer MRI data were reconstructed using a re-gridding method for non-cartesian acquisition.¹⁷ Receiver phase-offset and local phase inhomogeneity were corrected, as previously described.⁸ MRI VDP was quantified using a semi-automatic method which normalizes the volume of ventilation defects to the total thoracic cavity volume.¹ Temporal differences were evaluated using paired samples t-tests. Univariable relationships were evaluated using Pearson (r) correlations. Multivariable models were generated, using the backwards approach, to explain changes observed post-treatment. Results were considered statistically significant when the probability of making a type I error was less than 5% (P<.05).

Results

Table 1 provides demographic, pulmonary function, and imaging characteristics for 27 participants (mean age, 54±15 years). Forced expiratory volume in 1-second (FEV₁, P=.003), the ratio of FEV₁ to forced vital capacity (FEV₁/FVC, P<.001), MRI RBC:membrane (P=.047), and VDP (P=.004) were different between Week-0 and Week-6. Table 2 shows MRI RBC:membrane (β=-.458, P=.008) and VDP (β=-.395, P=.02) measured at Week-0 predicted the change in FeNO, while RBC:membrane (β=.517, P=.007) and the oscillometry measurement for airways reactance (β=-.452, P=.02) predicted the change in FEV₁/FVC. Figure 1 shows that, at Week-0, RBC:membrane correlated with FeNO (r=.56, P=.002), FEV₁ (r=-.45, P=.02), and FEV₁/FVC (r=-.60, P=.001).

Discussion

Using ¹²⁹Xe MRS, we identified that RBC:membrane, measured prior to treatment, was predictive of early response to ICS/LAMA/LABA treatment in moderate asthma. RBC:membrane correlated with markers of airway inflammation (FeNO) and obstruction (FEV₁/FVC) and was a significant predictor of improvement in these same measurements following 6-weeks of therapy. This may suggest RBC:membrane is indicative of alveolar-membrane gas-exchange abnormalities driven by upstream airway inflammation and obstruction, as previously shown in obstructive lung disease.⁹ In this way, ¹²⁹Xe MRS revealed novel vascular pathophysiology that may contribute to understanding the response to therapy in this group of patients.

Conclusion

For the first time, ¹²⁹Xe MRS measurements of pulmonary gas-exchange in the alveolar-membrane and red-blood-cells predicted early (6-week) treatment response to ICS/LAMA/LABA in moderate asthma patients. This finding helps frame a new paradigm for pulmonary vascular or gas-exchange treatable targets in asthma management.

Acknowledgements

Study 217438 is a Supported Collaborative Study with GSK.

References

1 Kirby, M. et al. Acad Radiol (2012).

2 Svenningsen, S. et al. Thorax (2014).

3 Svenningsen, S. et al. Am J Respir Crit Care Med (2018).

4 Eddy, R. L. et al. Am J Respir Crit Care Med (2020).

5 Svenningsen, S. et al. Eur Respir J (2016).

6 Svenningsen, S. et al. Chest (2020).

7 McIntosh, M. J. & Kooner, H. K. et al. Chest (2022).

8 Kaushik, S. S. et al. Magn Reson Med (2016).

9 Wang, Z. et al. Eur Respir J (2019).

10 Wang, J. M. et al. Thorax (2018).

11 Ash, S. Y. et al. Am J Respir Crit Care Med (2018).

12 McIntosh, M. J. et al. Am J Respir Crit Care Med (2023).

13 Parraga, G. et al. Invest Radiol (2007).

14 Svenningsen, S. et al. J Magn Reson Imaging (2013).

15 Matheson, A. M. et al. Radiology (2022).

16 Walker, T. G. & Happer, W. Rev Mod Phys (1997).

17 Robertson, S. H. et al. Concepts Magn Reson A (2015).