Ga-68 MAA Perfusion 4D-PET/CT Scanning Allows for Functional Lung Avoidance Using Conformal Radiation Therapy Planning

Abstract

Background:

Ga-68-macroaggregated albumin (⁶⁸Ga-perfusion) positron emission tomography/computed tomography (PET/CT) is a novel imaging technique for the assessment of functional lung volumes. The purpose of this study was to use this imaging technique for functional adaptation of definitive radiotherapy plans in patients with non-small cell lung cancer (NSCLC).

Methods:

This was a prospective clinical trial of patients with NSCLC who received definitive 3-dimensional (3D) conformal radiotherapy to 60 Gy in 30 fx and underwent pretreatment respiratory-gated (4-dimensional [4D]) perfusion PET/CT. The “perfused” lung volume was defined as all lung parenchyma taking up radiotracer, and the “well-perfused” lung volume was contoured using a visually adapted threshold of 30% maximum standardized uptake value (SUVmax). Alternate 3D conformal plans were subsequently created and optimized to avoid perfused and well-perfused lung volumes. Functional dose volumetrics were compared using mean lung dose (MLD), V5 (volume receiving 5 Gy or more), V10, V20, V30, V40, V50, and V60 parameters.

Results:

Fourteen consecutive patients had alternate radiotherapy plans created based on functional lung volumes. When considering the original treatment plan, the dose to perfused and well-perfused functional lung volumes was similar to that of the conventional anatomical lung volumes with an average MLD of 12.15, 12.67, and 12.11 Gy, respectively. Plans optimized for well-perfused lung improved functional V30, V40, V50, and V60 metrics (all P values <.05). The functional MLD of well-perfused lung was improved by a median of 0.86 Gy, P < .01. However, plans optimized for perfused lung only showed significant improvement in the functional V60 dose parameter (median 1.00%, P = .04) but at a detriment of a worse functional V5 (median 3.33%, P = .05).

Conclusions:

This study demonstrates proof of principle that 4D-perfusion PET/CT may enable functional lung avoidance during treatment planning of patients with NSCLC. Radiotherapy plans adapted to well-perfused but not perfused functional lung volumes allow for reduction in dose to functional lung using 3D conformal radiotherapy.

Keywords

perfusion scintigraphy positron emission tomography lung cancer functional imaging 4D imaging radiation therapy PET/CT

Introduction

Local treatment failures are still a major cause for the disappointing outcomes for patients with nonsmall-cell lung cancer (NSCLC) treated with radiotherapy. Locoregional failures still occur in up to 37% of patients¹ and is a major cause of the morbidity and mortality related to this disease. To minimize the risk of failure, the focus of the current international research is on intensification of radiotherapy treatment. Efforts to intensify radiotherapy are severely limited by the need to constrain dose to the surrounding normal lung in order to preserve function.² Unfortunately, acute lung injury secondary to radiation therapy (RT) in the form of pneumonitis is a potentially debilitating toxicity, sometimes leading to death of the patient. A recent meta-analysis suggests that symptomatic pneumonitis still occurs in 29.8% of patients and fatal pneumonitis in 1.9%.³ However, the currently used RT planning constraints that are designed to limit the risk of pneumonitis are based on evidence over a decade old.⁴ These constraints are based on population-based volumetric measurements of total irradiated lung irrespective of regional variation in function and do not account for individual variation in pulmonary physiology. Recent efforts to dose escalate without personalizing radiotherapy planning to the individual’s risk of pneumonitis have met with limited or no success.⁵ On the other hand, it has been estimated that tumor control probability (or likelihood of cure) for conventional radiotherapy could be improved by ∼50% (from 19.9% to 28.7%) by adaptively intensifying radiotherapy.⁶

Ventilation and perfusion (V/Q) imaging is an in vivo technique that measures regional lung function and may be used to individualize lung radiotherapy. Although both V/Q contribute to overall lung function, one does not necessarily match the other. Mismatched perfusion deficits may occur due to vascular disruption or pulmonary embolism, while reverse mismatch may occur due to severe obstructive airways disease⁷ or tumor-associated ventilatory obstruction. Assessment of lung perfusion (Q) is particularly relevant to radiation-induced lung damage as, along with pneumocytes, vascular endothelium is considered one of the most radiation-sensitive tissue in the lungs.⁸ Preferential sparing of pulmonary parenchyma with intact perfusion during radiotherapy is likely to reduce subsequent vascular injury and may ameliorate the risk of resultant toxicity from pneumonitis. Planar scintigraphy using ^99mTc-labeled macroaggregated albumin (MAA) is a long-established imaging standard for functional lung perfusion evaluation. Single positron emission computed tomography (SPECT) is a more modern functional assessment technique enabling 3-dimensional (3D) imaging,⁹ which has lead to improved sensitivity, specificity, and reproducibility.^10
-12 The advent of hybrid SPECT/CT devices further improved diagnostic accuracy by enabling anatomic characterization of scintigraphic abnormalities.¹³ Perfusion SPECT/CT has been demonstrated to improve functional lung avoidance during lung radiotherapy planning by several groups.^14
-16 However, at present it is unknown which specific threshold of lung perfusion constitutes the most clinically significant functional lung for radiotherapy avoidance. This is an area that warrants further clinical investigation.

Positron emission tomography/computed tomography (PET/CT) offers a unique opportunity to further improve the image quality of functional lung imaging owing to its superior sensitivity for detecting radioactive substances, higher spatial and temporal resolution, and commercial availability of respiratory-gated 4-dimensional (4D) acquisition systems.¹⁷ By substituting the conventional ^99mTc radionuclide with ⁶⁸Ga, a positron emitter, it is now possible to perform CT coregistered perfusion ⁶⁸Ga-MAA PET.¹⁸ We have previously reported that nongated 3D V/Q PET/CT has superior image quality and provides fully tomographic images with potential for better regional quantitation of lung function when compared to V/Q SPECT/CT in the context of pulmonary embolism.¹⁹ We have further improved this technique through the use of respiratory-gated (4D) acquisition, demonstrating that optimal image registration and spatial fidelity can be achieved using coregistration of specific phases of the 4D PET and 4D CT,²⁰ thereby reducing blurring caused by respiration motion and resultant artifact at the lung bases.

The purpose of this study is to investigate the utility of 4D perfusion PET/CT in facilitating functional avoidance of the lungs during radiotherapy planning. We aim to use the pretreatment perfusion PET/CT images of patients with NSCLC undergoing curative intent radiotherapy in order to simulate radiotherapy delivery based on avoidance of perfused lung. The central hypothesis of this study is that radiotherapy planning informed by perfusion PET/CT can be used to reduce dose to functional regions of lung.

Materials and Methods

The present work was part of observational prospective clinical trial (Universal Trial Number U1111-1138-4421) of patients undergoing curative intent radiotherapy with or without chemotherapy for NSCLC. This study received institutional review board approval from the Peter MacCallum Cancer Centre. All patients had pretreatment spirometry with testing for diffusing capacity of the lung for carbon monoxide (DLCO). As part of this trial, patients underwent pretreatment perfusion PET/CT. All scans were performed on a GE-Discovery 690 PET/CT scanner (GE Medical Systems Milwaukee, Wisconsin) in a default planning position using the radiotherapy palette and head rest with their arms raised. The patient's breathing trace was tracked using the Varian RPM respiratory tracking system (Varian Medical Systems, Palo Alto, California). The patients were instructed to breathe freely for the duration of the scans. No breathing training or coaching was used. A contemporaneous low-dose chest 4D-CT acquisition was performed using 140 kVp energy photons at 10 mA and a slice thickness of 5 mm. After intravenous cannulization, approximately 40 MBq ⁶⁸Ga-MAA was injected as a bolus. Immediately after administration of ⁶⁸Ga-MAA, a 3D List-mode respiratory-gated PET scan was acquired over the chest (2 bed positions, 5 minutes per bed position, and slice thickness 3.27 mm). The field of view for both 4D-PET and 4D-CT encompassed the entire lung fields. This acquisition was subsequently reconstructed as both a gated (4D) and an ungated scan. Phase-matched attenuation correction with 5 and 10 respiratory bins was used to reconstruct the 4D-PET/CT scan. The free-breathing PET acquisition was subsequently coregistered with the average (AVG) intensity projection of the 4D CT for purposes of treatment planning as described previously,²⁰ with subsequent PET thresholding applied to the time-AVG 4D PET images.

Radiotherapy Treatment Planning

All patients were planned to receive 60 Gy in 30 fractions of radiotherapy using 3D conformal techniques. Radiotherapy planning scans were performed with patients positioned supine with both arms elevated above the head. A respiratory-sorted 4-dimensional computed tomography (4DCT) data set was generated using the Philips Brilliance (Koninklijke Philips Electronics, Amsterdam, the Netherlands) CT scanner coupled with a Philips Bellows system or a Varian RPM system. The 4DCT scans were acquired in helical mode and binned into 10 phases for image reconstruction. The patients were imaged using 140 kVp, 3-mm slice thickness, 3-mm increment, and 0.44-s rotation time, and images were reconstructed with ∼3.5-mm³ voxel resolution (3-mm slice thickness × 1.0742-mm pixel spacing). From the respiratory-sorted imaging phases, AVG and maximum intensity projection (MIP) series were reconstructed. Target delineation was performed on an Elekta FocalSim (Medical Intelligence, Schwabmünchen Germany) workstation. Standardized lung window/level settings (1700/−300) were used in the MIP image series. An internal target volume was delineated from the MIP series, and a further isotropic expansion of 5-mm was used to generate the clinical target volume (CTV) and a further 10-mm isotropic expansion from CTV was used to create the planning target volume (PTV). The lung organ at risk (OAR) was defined as the volume of both lungs minus the volume of the GTV. Treatment planning was performed on the Elekta CMS/XiO V4.4 planning system blinded to the findings of the perfusion PET/CT. Doses to OARs were calculated using the AVG series. Typically a 3- to 4-field radiotherapy technique using 6 MV photons was used with effort made to avoid the contralateral unaffected lung and spare spinal cord, while ensuring the PTV was within −5% and +7% of the prescribed dose, per International Committee on Radiation Units and Measurements (ICRU) 62 recommendations. Dose constraints to OARs dose were as follows: spinal canal ≤45 Gy, mean lung dose (MLD) ≤20 Gy, V5 ≤60%, V20 ≤35%, and V30 ≤30%.

Functional Avoidance Radiotherapy Planning

Functional avoidance radiotherapy planning was performed on a data set created from rigidly registering 4D-PET/CT scan with the 4DCT radiotherapy planning scan. All Digital Imaging and Communications in Medicine (DICOM) data sets were imported into the iPlan treatment planning system (v4.5, BrainLAB; Feldkirchen, Germany). A single nuclear medicine physician (MSH) and a single radiation oncologist (SS) contoured functional lung volumes defined as either “perfused” or “well-perfused” lung using a visually adapted threshold technique. The perfused lung volume was defined as any perfused lung parenchyma, as defined by lung tissue with ⁶⁸Ga-MAA radiotracer uptake. The well-perfused lung was defined using an automated contour of lung parenchyma with a maximum standardized uptake value (SUVmax) threshold of 30% (ie, inclusion of the top 70% of voxels), which was then visually adjusted to exclude any central tracer clumping (Figure 1). Two functionally guided radiotherapy plans were then generated based on functional lung by a single radiotherapy planner (TD), a 3D conformal plan optimized to avoid perfused lung and a second 3D conformal plan optimized to avoid the well-perfused lung, respectively. Functional radiotherapy plans were limited to the use of the same or fewer numbers of beams as the original plan used for treatment. Dose volumetrics of the functional lung volumes (ie, calculating MLD and Vdose by excluding nonfunctioning lung) were compared to the “anatomical lung” volume as defined by the geometric lung OAR volume used for treatment. Each plan was optimized to the respective lung volume independently without being informed by the other available lung volumes.

Figure 1.

Hybrid ⁶⁸Ga-perfusion PET/CT imported into the iPlan planning system, demonstrating a patient with severe emphysema. Perfusion alone images is shown on the left panel, with coregistered PET/CT shown on the right panel. Anatomical (cyan contour), perfused lung (yellow contour), and well-perfused (red contour) lung volumes are displayed. A visually adapted standardized uptake value (SUV) thresholding technique was used to define well-perfused lung. PET/CT indicates positron emission tomography/computed tomography.

Statistical Analyses

The anatomical, perfused, and well-perfused lung volumes were compared using repeated measures analysis of variance. The following volumetric parameters were compared using Wilcoxon signed rank test: MLD, V5, V10, V20, V30, V40, V50, and V60. Median differences and their respective nonparametric confidence intervals were reported.

Results

Between December 2011 and September 2012, 14 consecutive patients with NSCLC were enrolled into this prospective study. The median (range) age of patients was 65 (47-90) years old. There were 5 patients with AJCC stage I disease, 1 patient with stage II disease, 7 patients with stage III disease, and 1 patient with stage IV disease (with resected brain metastasis). The most common tumor location was the RUL (n = 7). Patient characteristics and lung volumetrics are outlined in Table 1. The patient cohort consisted of exsmokers with preexisting airways disease, with a median (range) DLCO of 54% (27%-70%). The mean (95% CI) PTV volume was 400 cm³ (238-562 cm³). The mean volumes (95% CI) were 3722 cm³ (3016-4428 cm³), 3026 cm³ (2538-3514 cm³), and 1936 cm³ (1497-2376 cm³) for the anatomical, perfused, and well-perfused lung volumes, respectively. These volumes were significantly different on multiple comparison testing (P < .001). Figure 2A-D demonstrates a representative perfusion PET/CT in a patient with severe emphysema, the original plan used for treatment based on the anatomical lung, a plan optimized for perfused lung, and a plan optimized for well-perfused lung, respectively. All plans met the ICRU 62 dose prescription and target coverage goals.

Table 1.

Patient Characteristics.

Age	T	N	M	Stage	Tumor Location	Histology	% Pred. DLCO	PTV Volume, cm³	Anatomical Lung Volume, cm³	Perfused Lung Volume, cm³	Well-Perfused Lung Volume, cm³
61	4	0	0	IIIA	RLL	NSCLC NOS	70	469.9	4053.2	3855.5	2512.0
79	2	0	0	IB	RUL	SCC	56	437.2	4150.3	3300.7	2493.3
62	4	2	0	IIIB	RUL	SCC	60	812.9	5436.9	4452.2	3348.1
63	4	1	1b	IV	RUL	AdenoCa	52	422.5	4191.1	3586.5	1102.0
90	1	0	0	IA	LUL	SCC	43	140.2	3746.1	2484.4	1126.8
65	1	0	0	IA	RUL	SCC	45	95.6	4323.3	3858.4	3150.1
61	3	1	0	IIIA	LLL	Adeno	70	355.3	1588.5	1515.4	1396.7
73	4	2	0	IIIB	RUL	SCC	59	423.7	2780.5	2328.8	2288.5
65	1	2	0	IIIA	LUL	Adeno	39	342.8	2652.2	2690.3	1835.5
89	2	0	0	IIB	RUL	SCC	62	343.3	3279.7	2453.6	1787.5
47	2	2	0	IIIA	LUL	Adeno	44	297.3	2201.7	1991.3	1423.4
69	1	0	0	IA	RUL	NSCLC NOS	36	238.1	5258.6	3970.2	841.2
59	x	3	0	IIIB	^a	Large cell	64	1137.7	2869.8	2869.1	2187.0
67	1	0	0	IA	LLL	AdenoCa	27	87.2	5580.0	3010.0	1613.6

Abbreviations: adenoCa, adenocarcinoma; DLCO, diffusing capacity of the lung for carbon monoxide; LUL, left upper lobe; LLL, left lower lobe; NSCLC/NOS, nonsmall cell lung cancer; PTV, planning target volume; RUL, right upper lobe; RLL, right lower lobe; SCC, squamous cell carcinoma; T, Tumour; N, Lymph Node; M, Metastasis.

^aPatient with bulky bilateral mediastinal disease, no pulmonary parenchymal primary disease.

Figure 2.

Demonstrates planning methodology used in study. A, Perfusion positron emission tomography (PET) showing the PTV contoured in cyan, (B) the plan used for treatment with “perfused” and “well-perfused” lung contours demonstrating avoidance of contralateral anatomical lung, (C) an alternate plan optimized to perfused lung contours, and (D) an alternate plan optimized to well-perfused lung contours.

Three-dimensional conformal lung planning optimized to avoid regions of functional lung resulted in improved dose volumetric outcomes to functioning lung. Conventional planning techniques blinded to the findings of the perfusion PET/CT resulted in minimal or no significant changes to the functional lung dose volumetrics (Table 2). When alternate plans were optimized to the perfused lung volume, the V60 dose volume parameter improved by a median of 1% to the functional lung; however, this was at the expense of a worse V5 dose volume parameter, at a median detriment of 3.33% (Table 3). When optimizing the plan to the well-perfused lung volume, the median dose was significantly reduced to the functional lung for the MLD, V30, V40, V50, and V60 dose volume parameters (Table 4).

Table 2.

Evaluation of the Conventional Plan Optimized for Anatomical Lung, Comparing Conventional Dose Volumetrics Versus “Perfused” Lung and “Well-Perfused” Lung Dose Volumetrics.^a

DVH Parameter	Anatomical Lung, Median (Range)	Perfused Lung, Median (Range)	Well-Perfused Lung, Median (Range)	Anatomical Lung Minus Perfused Lung, Median Difference (95% CI)	P Value	Anatomical Lung Minus Well-Perfused Lung, Median Difference (95% CI)	P Value
Mean, Gy	11.2 (6-19.1)	11.4 (6.1-18.5)	11.4 (8.5-20.5)	0.41 (−0.11 to 0.82)	.194	−0.47 (−2.69 to 1.11)	.426
V5, %	35.8 (14.6-60.2)	35.7 (16.9-59.3)	36.6 (27.9-58.9)	−0.05 (−1.33 to 0.88)	.903	−1.50 (−5.60 to 2.22)	.391
V20, %	19 (9.5-31.4)	19.1 (9.5-30.3)	19.6 (10.7-34.1)	0.39 (−0.89 to 1.18)	.463	−1.39 (−5.56 to 2.61)	.426
V30, %	15.5 (8.7-26.3)	15 (8.5-25.2)	17 (8.4-29.8)	0.55 (−0.67 to 1.37)	.241	−1.03 (−4.94 to 2.18)	.502
V40, %	10.2 (4.2-22.8)	10.1 (3.9-21.5)	10.7 (4.1-26.6)	0.49 (−0.32 to 1.31)	.173	−0.54 (−4.06 to 1.24)	.626
V50, %	6.4 (2.2-18.9)	5.5 (1.6-16.5)	6.3 (2-18.7)	0.79 (0.12 to 1.35)	.011	−0.14 (−2.91 to 0.98)	.808
V60, %	2.1 (0.1-16.8)	1.8 (0-14.9)	2.1 (0.1-17)	0.38 (0.05 to 0.78)	.030	−0.1 (−1.87 to 0.55)	.529

Abbreviations: CI, confidence interval; DVH, dose–volume histogram.

^aStatistically significant findings highlighted in bold. Differences were calculated on a per patient basis.

Table 3.

Comparison of the Conventional Plan Versus the Alternate Plan Optimized for “Perfused Lung,” Evaluating “Perfused Lung” Dose Volumetrics.^a

Perfused Lung DVH Parameter	Conventional Plan, Median (Range)	Alternate Plan, Median (Range)	Conventional Plan Minus Alternate Plan, Median Difference (95% CI)	P Value
Mean, Gy	11.4 (6.1-18.5)	11.4 (6.9-17.6)	0.07 (−0.65 to 0.67)	.761
V5 (%)	35.7 (16.9-59.3)	37.6 (30.2-58.3)	−3.33 (−8.54 to −0.11)	.049
V20 (%)	19.1 (9.5-30.3)	20.3 (8.6-33)	−0.42 (−1.81 to 0.55)	.426
V30 (%)	15 (8.5-25.2)	14.1 (6.4-24.5)	1.07 (−0.38 to 2.11)	.119
V40 (%)	10.1 (3.9-21.5)	8.9 (4-18.6)	1.15 (−0.81 to 2.21)	.153
V50 (%)	5.5 (1.6-16.5)	5.3 (1.6-16.1)	1.27 (−0.12 to 2.75)	.119
V60 (%)	1.8 (0-14.9)	1.2 (0-12.9)	1.00 (0.02 to 1.92)	.042

Abbreviations: CI, confidence interval; DVH, dose–volume histogram.

^aStatistically significant findings highlighted in bold. Differences were calculated on a per patient basis.

Table 4.

Comparison of the Conventional Plan Versus the Alternate Plan Optimized for “Well-Perfused” Lung, Evaluating “Well-Perfused Lung” Dose Volumetrics.^a

Well-Perfused Lung DVH Parameter	Conventional Plan, Median (Range)	Alternate Plan, Median (Range)	Conventional Plan Minus Alternate Plan, Median Difference (95% CI)	P Value
Mean, Gy	11.4 (8.5-20.5)	10.6 (6.1-18.7)	0.86 (0.20 to 1.71)	.004
V5, %	36.6 (27.9-58.9)	32.6 (20.9-60.7)	−0.15 (−4.19 to 4.79)	.903
V20, %	19.6 (10.7-34.1)	19.4 (9.1-37)	0.46 (−0.41 to 1.37)	.235
V30, %	17 (8.4-29.8)	14.8 (6.3-25.6)	1.76 (0.28 to 3.50)	.030
V40, %	10.7 (4.1-26.6)	9.5 (4.4-19.1)	1.25 (0.10 to 3.53)	.042
V50, %	6.3 (2-18.7)	5.8 (1.7-14.6)	0.74 (0.11 to 2.38)	.020
V60, %	2.1 (0.1-17)	1.1 (0-9.8)	0.99 (0.11 to 3.81)	.011

Abbreviation: CI, confidence interval; DVH, dose–volume histogram.

^aStatistically significant findings highlighted in bold. Differences were calculated on a per patient basis.

Discussion

In this study, we propose 4D-⁶⁸Ga-perfusion PET/CT as a novel imaging modality to facilitate regional avoidance of functional lung during radiotherapy. We were able to demonstrate the significant benefits in functionally adaptive planning in terms of reduction in dose to functional lung. On visual evaluation, it was apparent that it was the distribution of perfusion across diseased lung that most influenced the capacity for functional lung avoidance, not the absolute size of perfused lung or the PTV. In this scenario, PET/CT has several inherent advantages in comparison to SPECT/CT imaging. The perfusion-PET/CT procedure image is acquired in a shorter time than a perfusion SPECT/CT scan, thus is more convenient for the patient, particularly in those with significant respiratory distress or discomfort related to their cancer or background lung disease. The PET/CT also yields images of higher quality due to superior counting sensitivity and resolution. The intrinsically quantitative nature of PET also facilitates more accurate estimates of perfused functional lung volume, particularly if acquired as a respiratory-gated (4D) study. Previous research by our group¹⁹ and others^21,22 has validated 3D-⁶⁸Ga-ventilation/perfusion PET/CT as an accurate imaging biomarker of pulmonary pathophysiology. Our experience with acquisition of 4D respiratory-gated PET imaging has been applied to further enhance the interpretation of perfusion PET/CT in the context of this study, through mitigation of respiratory-induced image artifact.

Loss of perfusion in response to radiotherapy has been previously documented through the use of older scintigraphic techniques. SPECT/CT imaging is an advance on planar V/Q scintigraphy that allows for in vivo assessment of regional lung perfusion. Post-irradiation, regional lung perfusion deficit is evident on SPECT imaging within 3 to 9 months after RT.^23,24 Lung perfusion seen in SPECT imaging is thought to be a surrogate for lung function based on phantom²⁵ and animal²⁶ experiments. The dose–effect relationship in perfused regions of lung appears consistent when detected by SPECT/CT in patients irradiated for lung cancer, breast cancer, and lymphoma.²⁷ A recent publication in NSCLC²⁸ found that the dose to perfused functional lung was a stronger predictor for toxicity (area under the curve [AUC] 0.76-0.98) than the dose to conventionally measured lung (AUC 0.62-0.86). However, SPECT/CT suffers from major limitations. The SPECT/CT is unable to account for subtle spatial and temporal changes in lung function, has poorer resolution than PET, and has limited capacity to account for breathing motion. Zhang et al,²⁹ the authors of the largest prospective series (spanning 14 years) of SPECT/CT, concluded that analysis of a dose–effect relationship in the lung using SPECT/CT was limited by inaccuracies including changes in patient anatomy over time and image registration errors from breathing. Given that the estimated radiation dose from perfusion PET to the patient is comparable to perfusion SPECT/CT scanning and lower than other blood flow assessments such as CT pulmonary angiography,³⁰ PET represents a logical new advance that may facilitate more accurate investigation of perfusion loss secondary to lung irradiation.

Selection of an appropriate threshold for what is considered functional lung is challenging in light of the variability in previously reported literature involving SPECT technology. It is further hampered by the lack of correlative validation with clinical outcomes of these previously reported functional lung volumes. There is presently no consensus in the optimal definition of what constitutes the functional perfused lung volume. As such our definitions of perfused were based on those used in previous studies,^14,15 and our definition of well-perfused lung was based on visual assessment and our own clinical experience suggesting that an SUVmax threshold of 30% (highest perfused 70%) represents a reproducible visual threshold defining apparent high functioning lung. A previous study of 16 patients by Shioyama et al using SPECT/CT³¹ optimized lung plans on the 50th and 90th percentile of functional lung and showed that the mean functional lung dose could be reduced by 2.2 and 4.2 Gy, respectively. McGuire et al³² used an iterative process to define the highest functioning lung in 5 patients and demonstrated that 4-beam intensity-modulated radiotherapy (IMRT) planned to avoid the highest functional lung could reduce the V20 and V30 to functional lung by a mean (±standard deviation) of 16.5% (±6.8%) and 6.1% (±9.2%), respectively, when compared to 9-beam conventional IMRT plans. However, a potential drawback of adaptively planning to only the highest 10% of perfused lung is the theoretical disadvantage of potential deposition of additional dose in regions of moderately perfused lung which may in fact contribute to the patient’s overall lung function.

On the other end of the spectrum, Christian et al¹⁴ defined functional lung volumes by threshold adjusting the SPECT perfusion image to match the volumes defined on CT. This approach would incorporate a large proportion of total available lung and may account for why there was no significant improvement in dose to functional lung found in adaptive planning for 5 of the 6 patients. The mean difference between conventional and functional lung measurements in that study was 1012 cm³. In comparison with our study, the mean difference between conventional volumes versus the perfused volumes (incorporating all radiotracer uptake in lung) was 696 cm³. Similar to Christian et al,¹⁴ our biologically adapted plans optimized to just the perfused lung volumes did not demonstrate significant improvements in dose to functional lung. However, adapting radiotherapy plans to the highest perfused 70% (30th centile) of lung in our study allowed a median reduction in MLD and V30 by 7.5% and 1.8%, respectively. This indicates that an appropriate selection of a threshold of lung perfusion may be necessary to demonstrate gains in functionally adaptive planning. Whether a selected threshold is clinically relevant is the subject of validation in future studies correlating clinical symptoms to functional dose volumetrics.

Conclusion

This is the first study incorporating the use of perfusion-PET in lung radiotherapy planning and demonstrating proof of principle that this technology can be used for functional lung avoidance. Our study demonstrated modest but statistically significant improvements in dose to functional lung volumes using functionally adapted radiotherapy plans without compromising target coverage. Importantly, this effect was only seen when adapting radiotherapy planning to the well-perfused areas of lung. In this study, an effort was made to constrain adaptive planning technique to be similar to the original technique used for treatment, using a 3D conformal technique with an equivalent number of beams as the original plan and avoidance of contralateral (functional) lung. Additionally, the study cohort consisted of a patient population with a modest MLD of 11.2 Gy. It may be that more advanced planning techniques, such as IMRT, volumetric modulated arc therapy, or incorporation of noncoplanar beams as is commonly employed in stereotactic radiotherapy, may allow for greater sparing of dose to functional lung. We suggest that further investigation into the use of 4D-⁶⁸Ga-perfusion PET/CT in functional lung avoidance is warranted using both larger patient cohorts and alternative radiotherapy planning techniques.

Footnotes

Abbreviations

Acknowledgments

We thank the invaluable contribution of Mr Peter Eu, for provision of radiopharmaceuticals used in this trial, and Mr Daniel Pham for guidance in the radiotherapy planning aspects.

Authors’ Note

SS is the principal investigator, responsible for oversight of trial and writing of the article. TD, TK, NH, and JC were responsible for radiotherapy planning, functional volume creation and contributed to writing of the article. DLB, MPM, FF, NP, DS, and MS were responsible for study design, conduct of trial, recruitment of patients onto trial, and contributed to writing of the article. RJH and MH were responsible for interpretation of functional imaging, contributed to design of the study, and writing of the article. MB performed the statistical analysis of this study. Australian and New Zealand Clinical Trials Registry, ACTRN12613000061730 (Universal Trial Number U1111-1138-4421).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr Shankar Siva has received National Health and Medical Research Council scholarship funding for this research, APP1038399. This trial was partially supported a Priority-drive Collaborative Cancer Research Scheme 2013 grant (APP 1060919).

References

Auperin

Le Pechoux

Rolland

. Meta-analysis of concomitant versus sequential radiochemotherapy in locally advanced non-small-cell lung cancer. J Clin Oncol. 2010;28(13):2181–2190. doi:10.1200/JCO.2009.26.2543.

Fay

Tan

Fisher

Mac Manus

Wirth

Ball

. Dose–volume histogram analysis as predictor of radiation pneumonitis in primary lung cancer patients treated with radiotherapy. Int J Radiat Oncol Biol Phys. 2005;61(5):1355–1363.

Palma

Senan

Tsujino

. Predicting radiation pneumonitis after chemoradiation therapy for lung cancer: an international individual patient data meta-analysis. Int J Radiat Oncol Biol Phys. 2013;85(2):444–450. doi:10.1016/j.ijrobp.2012.04.043.

Graham

Purdy

Emami

. Clinical dose-volume histogram analysis for pneumonitis after 3D treatment for non-small cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys. 1999;45(2):323–329.

Bradley

Paulus

Komaki

. 53rd Annual Meeting of the American Society of Radiation Oncology, Miami, FL, October 2-6, 2011.

Guckenberger

Kestin

Hope

. Is there a lower limit of pretreatment pulmonary function for safe and effective stereotactic body radiotherapy for early-stage non-small cell lung cancer? J Thorac Oncol. 2012;7(3):542–551.

Fan

Xia

Guan

Zhang

Liu

. Characteristic features of pulmonary function test, CT volume analysis and MR perfusion imaging in COPD patients with different HRCT phenotypes. Clin Respir J. 2014;8(1):45–54. doi:10.1111/crj.12033.

Hill

. Radiation effects on the respiratory system. Br J Radiol. 2005;27(Suppl):75–81. doi:10.1259/bjr/34124307.

Roach

Bailey

Harris

. Enhancing lung scintigraphy with single-photon emission computed tomography. Seminars in Nuclear Medicine. 2008;38:441–449. doi:10.1053/j.semnuclmed.2008.06.002.

10.

Jögi

Jonson

Ekberg

Bajc

. Ventilation–perfusion SPECT with 99mTc-DTPA versus Technegas: a head-to-head study in obstructive and nonobstructive disease. J Nucl Med. 2010;51(5):735–741.

11.

Gutte

Mortensen

Jensen

. Comparison of V/Q SPECT and planar V/Q lung scintigraphy in diagnosing acute pulmonary embolism. Nucl Med Commun. 2010;31(1):82–86.

12.

Roach

Bailey

Schembri

Thomas

PA.

Transition from planar to SPECT V/Q scintigraphy: Rationale, practicalities, and challenges. Seminars in Nuclear Medicine. 2010;40:397–407. doi: 10.1053/j.semnuclmed.2010.07.004.

13.

Roach

Gradinscak

Schembri

Bailey

Willowson

Bailey

. SPECT/CT in V/Q scanning. Seminars in Nuclear Medicine. 2010;40:455–466. doi:10.1053/j.semnuclmed.2010.07.005.

14.

Christian

Partridge

Nioutsikou

. The incorporation of SPECT functional lung imaging into inverse radiotherapy planning for non-small cell lung cancer. Radiother Oncol. 2005;77(3):271–277. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/j.radonc.2005.08.008.

15.

Lavrenkov

Christian

Partridge

. A potential to reduce pulmonary toxicity: the use of perfusion SPECT with IMRT for functional lung avoidance in radiotherapy of non-small cell lung cancer. Radiother Oncol. 2007;83(2):156–162. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/j.radonc.2007.04.005.

16.

Seppenwoolde

Engelsman

De Jaeger

. Optimizing radiation treatment plans for lung cancer using lung perfusion information. Radiother Oncol. 2002;63(2):165–177. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/S0167-8140(02)00075-0.

17.

Hicks

Hofman

. Is there still a role for SPECT–CT in oncology in the PET–CT era? Nat Rev Clin Oncol. 2012;9(12):712–20.

18.

Mathias

Green

. A convenient route to [68Ga]Ga-MAA for use as a particulate PET perfusion tracer. Appl Radiat Isot. 2008;66(12):1910–1912.

19.

Hofman

Beauregard

Barber

Neels

Hicks

. 68Ga PET/CT ventilation–perfusion imaging for pulmonary embolism: a pilot study with comparison to conventional scintigraphy. J Nucl Med. 2011;52(10):1513–1519.

20.

Callahan

Hofman

Siva

. High-resolution imaging of pulmonary ventilation and perfusion with 68Ga-VQ respiratory gated (4-D) PET/CT. Eur J Nucl Med Mol Imaging. 2014;41(2):343–349.

21.

Kotzerke

Andreeff

Wunderlich

Wiggermann

Zöphel

. Ventilation-perfusion-lungscintigraphy using PET and 68Ga-labeled radiopharmaceuticals. Nuklearmedizin. 2010;49(6):203–208.

22.

Ament

Maus

Reber

. PET lung ventilation/perfusion imaging using 68Ga aerosol (Galligas) and 68 Ga-labeled macroaggregated albumin. Recent Results Cancer Res. 2013;194:395–423.

23.

Woel

Munley

Hollis

. THE time course of radiation therapy-induced reductions in regional perfusion: a prospective study with >5 years of follow-up. Int J Radiat Oncol Biol Phys. 2002;52(1):58–67.

24.

Marks

Munley

Bentel

. Physical and biological predictors of changes in whole-lung function following thoracic irradiation. Int J Radiat Oncol Biol Phys. 1997;39(3):563–570.

25.

Osborne

Jaszczak

Coleman

Greer

Lischko

. In vivo regional quantitation of intrathoracic Tc-99m using SPECT: Concise communication. J Nucl Med. 1982;23(5):446–450.

26.

Osborne

Jaszczak

Greer

Lischoko

Coleman

. SPECT quantification of technetium-99m microspheres within the canine lung. J Comput Assist Tomogr. 1985;9(1):73–77.

27.

Seppenwoolde

Muller

Theuws

. Radiation dose–effect relations and local recovery in perfusion for patients with non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2000;47(3):681–690.

28.

Wang

Zhu

Sun

. Functional and biologic metrics for predicting radiation pneumonitis in locally advanced non-small cell lung cancer patients treated with chemoradiotherapy. Clin Transl Oncol. 2012;14(12):943–952. doi:10.1007/s12094-012-0890-3.

29.

Zhang

Zhou

. Radiation-induced reductions in regional lung perfusion: 0.1-12 year data from a prospective clinical study. Int J Radiat Oncol Biol Phys. 2010;76(2):425–432.

30.

Kotzerke

Andreeff

Wunderlich

. PET aerosol lung scintigraphy using Galligas. Eur J Nucl Med Mol Imaging. 2010;37(1):175–177.

31.

Shioyama

Jang

Liu

. Preserving functional lung using perfusion imaging and intensity-modulated radiation therapy for advanced-stage non–small cell lung cancer. Int J Radiat Oncol Biol Phys. 2007;68(5):1349–1358. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/j.ijrobp.2007.02.015.

32.

McGuire

Marks

Yin

Das

. A methodology for selecting the beam arrangement to reduce the intensity-modulated radiation therapy (IMRT) dose to the SPECT-defined functioning lung. Phys Med Biol. 2010;55(2):403–416.