Radiomic Feature Spatially Characterizes the Transition of Normal Lung Parenchyma to Small Airways Disease in COPD

1. Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of death and healthcare burden in the United States and worldwide. Accounting for over 3 million deaths globally in 2015 (“Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990–2015,” 2017), this disease is expected to rise in prevalence as the world population ages (Mannino & Buist, 2007). COPD is understood to be a complex heterogeneous disease presenting clinically diverse phenotypes (Barker & Brightling, 2013; Han et al., 2010). Major causes of airflow obstruction are attributed to chronic bronchiolitis, a.k.a small airways disease (SAD), and emphysema. Although SAD and emphysema are treated as separate COPD subtypes, studies have shown strong quantitative evidence that SAD exists as an intermediate state between healthy lung tissue and emphysema—i.e., irreversible lung damage—in COPD pathogenesis (Jennifer L. Boes et al., 2015; Labaki et al., 2019; McDonough et al., 2011). At present, little has been done to better quantify the onset of SAD from healthy lung parenchyma.

The Parametric Response Map (PRM) is a CT-based voxel-wise computational technique that can identify and quantify functional small airways disease (fSAD; indirect measure of SAD) even in the presence of emphysema (Galbán et al., 2012). The percent volume of PRM-derived fSAD (PRM^fSAD), i.e., the amount of fSAD in the lungs, has improved COPD phenotyping and the prediction of spirometric decline in subjects at risk of COPD (Bhatt et al., 2016). To determine the value of spatial features from each PRM classification, we developed topological PRM (tPRM) as an extension of the PRM algorithm (Hoff et al., 2017). These radiographic tPRM readouts were shown to improve upon commonly used whole-lung PRM measures with respect to COPD characterization, and correlate to structural changes in lung tissue samples from subjects diagnosed with bronchiolitis obliterans (Ram et al., 2022). Of interest is the Euler-Poincaré Characteristic (χ); positive values of this topological feature are associated with the formation of pockets of like tissue (e.g., pockets of fSAD) (Hoff et al., 2017). χ, when generated from PRM^fSAD (χ^fSAD), may serve to detect lung parenchymal transition from healthy to emphysema through SAD.

The transition from normal parenchyma to SAD, and the relationship of this transition to emphysema, is the focus of this study. We investigated χ^fSAD as an index of COPD severity in baseline (Phase 1) data from COPDGene (Regan et al., 2011) and demonstrated the utility of our methodology to assess disease phenotype and transition of fSAD from normal lung parenchyma.

2. Materials and Methods

3. Results

3.2 Topological Readouts of PRM

Presented in Figure 1 is a case with elevated fSAD (PRM^fSAD = 40%). Representative coronal slices of the expiration CT scan and PRM^fSAD, overlayed on CT scan, are provided. To illustrate the dependence of V and χ on the arrangement of PRM^fSAD, we have included V^fSAD and χ^fSAD maps indicating regions with low (cyan box) and high (magenta box) V^fSAD. As expected, V^fSAD (Figure 1C) is dependent on the amount of fSAD (yellow voxels in Figure 1B). Averaged over the lungs, V^fSAD is proportional to the percent volume of PRM^fSAD by a factor of 100. However, χ^fSAD > 0 (cyan box in Figure 1D) corresponds to the formation of fSAD pockets (cyan box Figure 1B), whereas χ^fSAD < 0 (cyan box in Figure 1D) is the consolidation of these pockets into a mesh with holes (magenta box in Figure 1B).

• Download figure
• Open in new tab

Figure 1:

Illustration of Volume Density (V) and Euler-Poincaré Characteristic (χ) for PRM^fSAD. Presented are representative coronal slices for the (A) expiratory CT scan with associated (B) PRM^fSAD overlay (yellow). Included are the (C) volume density and (D) Euler-Poincaré Characteristic of PRM^fSAD. Magenta box indicates a lung region with elevated V^fSAD and negative χ^fSAD. Cyan box indicates a lung region with minimal VfSAD and positive χ^fSAD. The subject is a GOLD 3 female (51-55 years old) with FEV1% predicted of 32% and percent volume of PRM^fSAD of 40%.

The volume density of PRM^Norm and PRM^fSAD demonstrated an inverse relationship with increasing COPD severity (Figure 2A), consistent with previous work. A similar inverse relationship was observed for χ of both normal lung and fSAD (χ^Norm and χ^fSAD). Values of χ^Norm and χ^fSAD were found to flip about zero (e.g., χ^fSAD changes from positive to negative values) from GOLD 2 to GOLD 4 (Figure 2B). As provided in Supplemental Table 1, χ^Norm and χ^fSAD cases diagnosed as GOLD 2. For those with severe COPD, i.e., GOLD 4, χ^Norm and χ^fSAD were 0.0039 (0.0055) and −0.0036 (0.0048), respectively. Mean values of χ^Emph and χ^PD were found to be positive and similar across GOLD status (Supplemental Table 1).

• Download figure
• Open in new tab

Figure 2:

Boxplots for Volume Density (V) and Euler-Poincaré Characteristic (χ) of PRM^Norm and PRM^fSAD across all GOLD stages, “at-risk” (GOLD 0) and PRISm. Plots of (A) V and (B) χ are provided for PRM^Norm (green) and PRM^fSAD (yellow). Box plots were computed following standard protocol for box and whiskers. Box lines determined by lower quartile (Q1), middle quartile / median (Q2), and upper quartile (Q3). Whiskers are drawn out to Q1 - 1.5 x IQR and Q3 + 1.5 x IQR for lower and upper limits respectively. IQR = Q3-Q1. Outliers are defined as points beyond the given upper and lower limits and illustrated as black points with a random bounded horizontal perturbation beyond box whiskers.

We further evaluated the relationship of PRM^Norm and PRM^fSAD with respect to V (Figure 3A) and χ (Figure 3B). Both V and χ demonstrated strong correlations between Norm and fSAD (ρ = −0.666, p < 0.001 and ρ = −0.745, p < 0.001, respectively) over the full study cohort. Here the GOLD stage is represented by different colors and the relative amount of emphysema, quantified by V^Emph, by size of the bubbles. As observed in Figure 3A, V^Norm versus V^fSAD had more scatter in the data compared to χ^Norm versus χ^fSAD (Figure 3B). Cases with elevated emphysema (V^Emph) demonstrated lower V^fSAD and χ^Norm > 0, but demonstrated positive and negative values of χ^fSAD (Figure 3B).

• Download figure
• Open in new tab

Figure 3:

Scatter plots of all study sample participants for (A) V^Norm versus V^fSAD and (B) χ^Norm versus χ^fSAD. Individual points are color coded based on COPD classifications. The size of the points indicates the amount of emphysema as measured by the volume density of PRM^Emph (V^Emph).

3.4 Dependence Between Topologies of PRM^fSAD and PRM^Emph

As expected from the multivariable analysis, V^fSAD was strongly correlated to V^Emph (r = 0.845, p < 0.001). Nevertheless, only a weak correlation was observed between χ^fSAD and χ^Emph (ρ = −0.155, p < 0.001). As the topologies of PRM were determined as averages over the whole lungs, we provide a case study illustrating the relationship between V and χ of PRM^fSAD and PRM^Emph at the local level. Presented in Figure 4 are the profiles of V and χ of PRM^fSAD and PRM^Emph from a region of the right lung with elevated and reduced V^Emph (orange circle and star, respectively, in Figure 4A and C). The case is a female subject (46-50 years old) diagnosed with GOLD 4 COPD. The subject was found to have over the lungs high levels of V^fSAD (0.37) with relatively elevated V^Emph (0.1). Mean values of χ were 0.008 and −0.009 for PRM^Emph and PRM^fSAD, respectively. As seen in Figure 4C, V^fSAD increased while V^Emph decreased further from lung with the highest level of V^Emph (∼0.6 at orange circle in Figure 4A and C). At approximately 1.8cm, volume densities between PRM^fSAD and PRM^Emph transitioned. Inversely, χ^fSAD was found to increase with decreasing χ^Emph with transition occurring at ∼1.2cm.

• Download figure
• Open in new tab

Figure 4:

Case study demonstrating the spatial relationship between the topologies of PRM^fSAD and PRM^Emph. The case is a female subject (46-50 years old) diagnosed with GOLD 4 COPD. Single axial slice from (A) spatially aligned CT scan acquired at full inflation with corresponding (B) slices from V and χ of PRM^fSAD and PRM^Emph. (C) Topology values were plotted along the dashed line on the CT slice, starting from circle to star. Lines on plot were color coded to match PRM classification (red signifies PRM^Emph and yellow signifies PRM^fSAD). Solid and dashed lines indicate V (left y-axis) and χ (right y-axis).

4. Discussion

The topological parametric response map is an extension of the well-established PRM method, a powerful quantitative imaging marker (Galbán et al., 2012). In this study, we have demonstrated that inclusion of topological features, in this case the Euler-Poincaré Characteristic (χ), improved characterization and interpretation of fSAD in COPD as a complimentary readout of volume density (V), which is equivalent to traditional percent volume of PRM classifications (Hoff et al., 2017). This study also evaluated the role of PRM-defined normal parenchyma (PRM^Norm) and fSAD (PRM^fSAD) as lone indicators of COPD severity. We observed distinct patterns in topological metrics with respect to GOLD grades and identified a complete inversion in topology, characterized by Euler-Poincaré Characteristic χ, between normal lung and fSAD, in mid-to-late stages of COPD. We also found statistically significant associations of spirometric measures and emphysema with V and χ of PRM^Norm and PRM^fSAD.

Our study builds on previous work by Hoff et al (Hoff et al., 2017) on tPRM characterization in COPD. This study used a much smaller population (n = 88) to demonstrate the trends of all four topological features (volume density, surface area, mean curvature and Euler-Poincaré Characteristic) with increasing COPD severity (Hoff et al., 2017). Limited in statistical power, it instead focused on the surface area of fSAD. Access to a notably larger population (n = 8,956) allowed us in the current study to evaluate the volume density (V) and Euler-Poincaré Characteristic (χ) of PRM^Norm and PRM^fSAD and relate our findings to the field’s current understanding of COPD progression; i.e., normal parenchyma transitions to emphysema through SAD.

A key finding of this study is the ability to quantify parenchymal lung health, based not only on the extent but also on the arrangement of local lung abnormalities, i.e., fSAD. This is rooted in the concept that the lungs are healthy (i.e., PRM^Norm) and COPD progresses through SAD (i.e., PRM^fSAD), which is an intermediate between normal and emphysematous lung tissue. The nature of this transition suggests χ may be capturing a fundamental mechanism in the emergence of fSAD. Based on our observation, fSAD appears to develop as distinct pockets within healthy lung tissue as depicted in cyan box in Figure 1B. With increasing COPD severity, fSAD pockets coalesce to a mesh, which is represented by decreasing values in χ^fSAD (magenta box in Figure 1B). On a whole lung level, this transition occurs on average from GOLD stages 2 to 4. By quantifying the amount and arrangement of normal and fSAD parenchyma, one can obtain a full assessment of the COPD lung. As fSAD is an intermediate between healthy lung and emphysema, increasing levels of emphysema have a direct effect on V and χ of fSAD. This is observed in Figure 3, where increasing values of V^Emph resulted in a drop in V^fSAD and increase in χ^fSAD. These trends were reflected in our multivariable model for V^Emph as well (Table 2).

We postulate that the transition observed between χ^Norm and χ^fSAD (Figures 2 and 3) should be observed for χ^fSAD and χ^Emph. Using mean values of χ over the lungs, χ^Emph was found to be relatively stable, generating positive values across GOLD (Supplemental Table 2). In a recent study, Bhatt and colleagues evaluated a CT readout, referred to as the mean Jacobian determinant of normal voxels, at varying distances from emphysematous tissue (Bhatt et al., 2017). When measured at a distance of 2mm from CT voxels designated emphysema (i.e., voxel HU <-950HU), this CT-based readout was found to be predictive of spirometric decline. We applied a similar strategy to determine if a relationship between χ^fSAD and χ^Emph is observed at the local level (Figure 4). Our spatial analysis of a single case clearly demonstrates a transition in topologies of PRM^fSAD and PRM^Emph. Although the readouts reported by Bhatt and colleagues lacked quantification of SAD, there is clear agreement that lung tissue along the periphery of emphysematous tissue provides potential insight into COPD progression. Additional evidence is from the seminal work by McDonough and colleagues (McDonough et al., 2011). In this study they provided pathological evidence demonstrating the role of SAD in COPD progression. Using high resolution (∼10 μm) microCT to analyze frozen lung samples from lung transplant recipients with end-stage COPD, they found that widespread narrowing and destruction of the smaller airways (i.e., SAD) occurred before emphysematous lesions became large enough to be visible on standard CT imaging. They concluded that SAD might serve as an emphysema precursor. We suspect that the transition of χ^fSAD and χ^Emph may be capturing what was observed using microCT of lung explants.

We acknowledge several notable limitations. COPDGene comprises over 20 study sites, making scanner variation and reconstruction kernel inconsistency inevitable. Sensitivity of PRM to scanner variability was addressed previously, and although effort was made to apply PRM only to soft kernels, variability in scanner type was unavoidable. However, we included scanner vendor in our multivariable regressions and found that it did not significantly confound models. Another limitation is variation in levels of inspiration and expiration during CT acquisition. Earlier work demonstrated that even small perturbations from functional residual capacity (FRC) have an observable effect on threshold-based techniques such as PRM (J. L. Boes et al., 2015). To limit this, we implemented QC that excluded participants based on erroneous volume changes or strong discordance with correlation between PRM^Norm and FEV₁% predicted.

5. Conclusions

In summary, we have demonstrated that topological features, V and χ, are able to enhance the sensitivity of PRM classifications, notably Norm and fSAD, to COPD severity. This data supports the concept that as pockets of small airways disease coalesce, surrounding normal tissue is lost. These pockets of fSAD are correlated with increasing presence of emphysema, independent of the amount of fSAD present. We further demonstrated at the local level that χ^fSAD and χ^Emph correlate, which may be explained by McDonough and colleagues’ observations using microCT. Our study provides a unique strategy to detect subtle changes in lung parenchyma that may progress to emphysema. This approach to monitoring extent and arrangement of Norm and fSAD may offer insight into COPD phenotypes and provide prognostic information that has relevance in clinical care and future clinical trials. Better computational strategies must be developed before these topologies can be used to detect local emphysema onset.

Funding

This work was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health Grants R01 HL139690 and R01 HL150023 and by the National Heart, Lung, and Blood Institute of the National Institutes of Health Grants U01 HL089897 and U01 HL089856, which support the COPDGene study. The COPDGene study (NCT00608764) is also supported by the COPD Foundation through contributions made to an Industry Advisory Committee comprised of AstraZeneca, Bayer Pharmaceuticals, Boehringer-Ingelheim, Genentech, GlaxoSmithKline, Novartis, Pfizer and Sunovion.

Conflict of Interest Statement

W.W.L. reports personal fees from Konica Minolta and Continuing Education Alliance. B.A.H. and C.J.G. are co-inventors and patent holders of tPRM, which the University of Michigan has licensed to Imbio, LLC. C.J.G. is co-inventor and patent holder of PRM, which the University of Michigan has licensed to Imbio, LLC. B.A.H. and C.J.G. have financial interest in Imbio, LLC. C.R.H. is employed by Imbio, LLC. D.A.L. reports funds paid to the institution from the NIH, and Boehringer Ingelheim. M.K.H. reports personal fees from GlaxoSmithKline, AstraZeneca, Boehringer Ingelheim, Cipla, Chiesi, Novartis, Pulmonx, Teva, Verona, Merck, Mylan, Sanofi, DevPro, Aerogen, Polarian, Regeneron, Amgen, UpToDate, Altesa Biopharma, Medscape, NACE, MDBriefcase and Integrity. She has received either in kind research support or funds paid to the institution from the NIH, Novartis, Sunovion, Nuvaira, Sanofi, AstraZeneca, Boehringer Ingelheim, Gala Therapeutics, Biodesix, the COPD Foundation and the American Lung Association. She has participated in Data Safety Monitoring Boards for Novartis and Medtronic with funds paid to the institution. She has received stock options from Meissa Vaccines and Altesa Biopharma. For the remaining authors none were declared.

Administrative Center

James D. Crapo, MD (PI); Edwin K. Silverman, MD, PhD (PI); Barry J. Make, MD; Elizabeth A. Regan, MD, PhD

Genetic Analysis Center

Terri Beaty, PhD; Ferdouse Begum, PhD; Peter J. Castaldi, MD, MSc; Michael Cho, MD; Dawn L. DeMeo, MD, MPH; Adel R. Boueiz, MD; Marilyn G. Foreman, MD, MS; Eitan Halper-Stromberg; Lystra P. Hayden, MD, MMSc; Craig P. Hersh, MD, MPH; Jacqueline Hetmanski, MS, MPH; Brian D. Hobbs, MD; John E. Hokanson, MPH, PhD; Nan Laird, PhD; Christoph Lange, PhD; Sharon M. Lutz, PhD; Merry-Lynn McDonald, PhD; Margaret M. Parker, PhD; Dmitry Prokopenko, Ph.D; Dandi Qiao, PhD; Elizabeth A. Regan, MD, PhD; Phuwanat Sakornsakolpat, MD; Edwin K. Silverman, MD, PhD; Emily S. Wan, MD; Sungho Won, PhD

Imaging Center

Juan Pablo Centeno; Jean-Paul Charbonnier, PhD; Harvey O. Coxson, PhD; Craig J. Galban, PhD; MeiLan K. Han, MD, MS; Eric A. Hoffman, Stephen Humphries, PhD; Francine L. Jacobson, MD, MPH; Philip F. Judy, PhD; Ella A. Kazerooni, MD; Alex Kluiber; David A. Lynch, MB; Pietro Nardelli, PhD; John D. Newell, Jr., MD; Aleena Notary; Andrea Oh, MD; Elizabeth A. Regan, MD, PhD; James C. Ross, PhD; Raul San Jose Estepar, PhD; Joyce Schroeder, MD; Jered Sieren; Berend C. Stoel, PhD; Juerg Tschirren, PhD; Edwin Van Beek, MD, PhD; Bram van Ginneken, PhD; Eva van Rikxoort, PhD; Gonzalo Vegas Sanchez-Ferrero, PhD; Lucas Veitel; George R. Washko, MD; Carla G. Wilson, MS;

PFT QA Center, Salt Lake City, UT

Robert Jensen, PhD

Data Coordinating Center and Biostatistics, National Jewish Health, Denver, CO

Douglas Everett, PhD; Jim Crooks, PhD; Katherine Pratte, PhD; Matt Strand, PhD; Carla G. Wilson, MS

Epidemiology Core, University of Colorado Anschutz Medical Campus, Aurora, CO

John E. Hokanson, MPH, PhD; Gregory Kinney, MPH, PhD; Sharon M. Lutz, PhD; Kendra A. Young, PhD

Mortality Adjudication Core

Surya P. Bhatt, MD; Jessica Bon, MD; Alejandro A. Diaz, MD, MPH; MeiLan K. Han, MD, MS; Barry Make, MD; Susan Murray, ScD; Elizabeth Regan, MD; Xavier Soler, MD; Carla G. Wilson, MS

Biomarker Core

Russell P. Bowler, MD, PhD; Katerina Kechris, PhD; Farnoush Banaei-Kashani, Ph.D

We would also like to acknowledge our copy editor Lee Olsen for her assistance in preparing this manuscript.