Synopsis

Radiation induced lung disease (RILD) can occur following thoracic irradiation, chemotherapy and immunotherapy in lung cancer patients, manifesting as inflammation, radiation pneumonitis or fibrosis. Oxygen Enhanced (OE)-MRI provides spatial information on oxygen delivery to lung tissue. Five non-small cell lung cancer patients underwent OE-MRI pre and post radiotherapy. Statistically significant changes were measured between pre and post radiotherapy scans in both the contralateral and ipsilateral lungs. These findings indicate that OE-MRI can assess changes in lung function following radiotherapy, with potential application for personalised therapy and reduction of toxicity.

Introduction

Radiation induced lung disease (RILD) can occur following thoracic irradiation and chemotherapy during lung cancer treatment and presents as inflammation, radiation pneumonitis and fibrosis. The volume of lung irradiated and the delivered radiation dose (both of which depend on tumour location, size and stage) can influence the degree of injury¹. Methods used to image lung function and structure in healthcare either contribute additional radiation dose or have limited availability². The advent of hybrid MR-Linac systems further motivates the development of improved MRI methods for assessing radiation damage during the repeat visits of fractionated radiotherapy treatment (RT)².

Oxygen enhanced (OE)-MRI is a non-ionising and non-invasive method providing spatial information on oxygen delivery to lung tissue³. In OE-MRI, inhaled oxygen dissolves in the blood plasma and interstitial spaces of the lung, increasing the spin-lattice relaxation rate (R₁) in proportion to the local change in partial pressure of oxygen (∆PO₂)⁴. This study is the first to investigate the utility of dynamic OE-MRI to measure changes in lung function following radiotherapy in lung cancer patients.

Methods

Five patients with non-small cell lung cancer (NSCLC) underwent dynamic OE-MRI imaging pre-radiotherapy and had a second scan at day (19±4 days). Total doses of 45±8 Gy were administered by the time of the second scan (labelled post RT below), as part of sequential chemo-radiotherapy. Radiotherapy prescribed dose plans were 55 Gy in 20 (N=3) or 60-66 Gy in 30-33 (N=2) fractions. All MRI data were acquired free-breathing on a 1.5 T MR scanner (Philips Achieva, Philips Medical Systems, Best, Netherlands) using Q Body (OE-MRI) and Sense XL Torso coils (anatomical scans). All participants provided written informed consent and the study was conducted following ethical approval.

R₁ was calculated using a series of coronal inversion recovery (IR) turbo field echo acquisitions (TR 2.1 ms, TE 0.496 ms, 6° flip angle, number of excitations = 5, TI = 10, 50, 300, 1100, 2000, 5000 ms) using a non-selective inversion and centric k-space ordering. Dynamic images were acquired with a TI of 1100 ms and temporal resolution of 10 s, enabling full magnetisation recovery between volumes. Twelve measurements were acquired while breathing medical air (21% oxygen; (15 l/min) to determine native R₁, before switching to 100% for 66 measurements and then 12 final measurements on medical air. Gases were delivered via a non-rebreathing Hudson mask. Maps of baseline T₁, maximum changes in PO₂ from 21-100% oxygen (ΔPO_2max), derived from fitting an exponential uptake function to the ΔPO₂(t), and the maximum percentage signal difference between 21-100% oxygen (max%diff) were generated³. Oxygen relaxivity constant of r = 2.49x10-4 s-1 mmHg-1 was used to calculate ΔPO2⁵.

Lung volumes were manually segmented on the pre and post radiotherapy images for both lungs in Matlab R2014a (Mathworks, USA), excluding the gross tumour volume (GTV; outlined by an experienced radiation oncologist on a reference radiotherapy planning scan) and large vessels. Lung volumes were propagated to native (baseline) T1, ΔPO_2max and max%diff maps and median parameter values were quantified per lung. Only enhancing voxels were included. Voxels with T₁ calculation fitting errors were discarded from the analysis. A two-tailed paired t-test was used to test for significance between pre and post radiotherapy measurements.

Results

Figure 1 shows example baseline T₁ and ΔPO_2max maps pre and post radiotherapy in a patient with a T3N2M0 tumour in the right lung (tumour not outlined).

The median baseline T₁, ΔPO_2max and max%diff results for the group of 5 patients for the ipsilateral (tumour containing) and contralateral lungs, pre and post radiotherapy are shown in table 1 and figures 2.a-f respectively.

Statistically significant reduction in ΔPO2max was measured following radiotherapy in the ipsilateral and contralateral lungs (p=0.03 for both). No other changes reached statistical significance.

Discussion

Conclusion

Acknowledgements

References

1. Yo Choi et al, Effects of Radiation Therapy on the Lung: Radiologic Appearances and Differential Diagnosis. RadioGraphics 2004; 24:985–99.
2. Bainbridge, H et al. Magnetic resonance imaging in precision radiation therapy for lung cancer. Transl Lung Cancer Res. 2017 Dec; 6(6): 689–707.
3. Martini, K., Gygax, C. M., Benden, C., Morgan, A. R., Parker, G. J. M., & Frauenfelder, T. (2018). Volumetric dynamic oxygen-enhanced MRI (OE-MRI): comparison with CT Brody score and lung function in cystic fibrosis patients. European Radiology, 28(10), 4037–4047.
4. Morgan, A. R., Parker, G. J. M., Roberts, C., Buonaccorsi, G. A., Maguire, N. C., Hubbard Cristinacce, P. L., Nordenmark, L. H. (2014). Feasibility assessment of using oxygen-enhanced magnetic resonance imaging for evaluating the effect of pharmacological treatment in COPD. European Journal of Radiology, 83(11), 2093–2101.
5. Zaharchuk, G., Busse, R. F., Rosenthal, G., Manley, G. T., Glenn, O. A., & Dillon, W. P. (2006). Noninvasive oxygen partial pressure measurement of human body fluids in vivo using magnetic resonance imaging. Academic Radiology, 13(8), 1016–1024.
6. Mirsadraee S et al. T1 characteristics of interstitial pulmonary fibrosis on 3T MRI – a predictor of early interstitial change? Quant Imaging Med Surg 2016;6(1):42-49.