Imaging of Pediatric Pulmonary Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper

Abstract

Pediatric pulmonary malignancy can be primary or metastatic, with the latter being by far the more common. With a few exceptions, there are no well-established evidence-based guidelines for imaging pediatric pulmonary malignancies, although, computed tomography (CT) is used in almost all cases. The aim of this article is to provide general imaging guidelines for pediatric pulmonary malignancies, including minimum standards for cross-sectional imaging techniques and specific imaging recommendations for select entities.

INTRODUCTION:

Among pediatric thoracic malignancies, metastatic disease and primary mediastinal tumors, such as lymphoma and neuroblastoma, are most common . These entities are discussed in detail in separate articles in this series; therefore, this article will address pediatric pulmonary malignacies.

Metastatic disease from an extracranial solid neoplasms is far more frequent than primary pulmonary malignancy in children, and mainly associated with Wilms tumor, sarcomas, and hepatoblastoma ¹. Primary pediatric pulmonary cancers are rare with little in the literature on this topic. The largest series was published by Hancock et al., comprising 383 total cases ², the majority of which (76%) were malignant. Yu et al. published a review encompassing ninety years of experience at a single institution and identified 14 individual pathologic types of pediatric primary lung tumors, the most common (overall amounting to 50% of cases) being pleuropulmonary blastoma (PPB) (15%), inflammatory myofibroblastic tumor (17.5%), and carcinoid tumor (20%).

The clinical presentation for primary malignant pulmonary tumors in children is dependent on lesion size and location, and its nonspecificity may result in delayed diagnosis. Patients manifest cough (35%), pneumonitis (23%), fever (18%), respiratory distress (12%), and hemoptysis (11%). A minority of children with pulmonary malignant lesions are asymptomatic (6%)³.

INITIAL WORK-UP:

In 2008, Weldon and Shamberger proposed an approach to identify pediatric pulmonary tumors based on symptoms and most likely non-neoplastic etiologies³. At presentation, the authors recommend chest radiography (CXR) in addition to laboratory tests and treatment directed towards the most likely disease. If symptoms persist respiratory consultation and a chest computed tomography (CT) examination are indicated. However, this recommendation is limited as a significant portion of pediatric pulmonary tumors are discovered unexpectedly.

STAGING:

Currently, there are no dedicated systems to stage pediatric pulmonary malignancies as a group. The adult American Joint Committee on Cancer (AJCC) TNM staging system may be applied to carcinoid tumors and bronchogenic carcinomas, but lacks validation for children; this method, based on both imaging and bronchoscopic findings, evaluates the tumor size, local invasion, and secondary features, such as post-obstructive pneumonitis. Regional nodal involvement may be assessed and documented with the International Association for the Study of Lung Cancer (IASLC) lymph node station map ⁴. The advantages and disadvantages of imaging modalities available for evaluating pediatric pulmonary malignancies are summarized in Table 2.

Table 2:

Chest CT Protocol for Pediatric Pulmonary Malignancies

Series Name	Coverage	Reconstructed Slice Thickness	Contrast Phase (if applicable)	Reformat plane/types/reconstruction kernel	Comment(s)
Chest Axial Lungs	Top of lungs through bottom of the lungs	1-3 mm	Mixed pulmonary arteriographic and venous phase	Axial plane Lung Kernel	Unless a contraindication exists, iodinated contrast should be administered if evaluating primary tumor; contrast unnecessary if only assessing for lung metastates
Chest axial Soft tissue	Top of lungs through bottom of the lungs	3-5 mm		Axial plane Soft tissue kernel
Chest coronal Reformat	Anterior skin through posterior skin	2-4 mm		Coronal plane Soft tissue kernel
Chest MIPs	Top of lungs through bottom of the lungs	5-15 mm		Axial or coronal plane Lung Kernel	Best series for depicting small pulmonary nodules

MIPS = Maximum intensity projection

THORACIC IMAGING FOR STAGING (Table 1):

Table 1:

Advantages and Disadvantages of Individual Imaging Modalities for the Evaluation of Pediatric Pulmonary Malignancies

Procedure Name	Timepoint(S)	Advantage(s)	Disadvantage(s)
Chest Radiograph	Initial	Quick Low Cost Widely available	Nonspecific findings May miss small pulmonary lesions May miss or incompletely demonstrate mediastinal involvement Requires radiation
CT	Diagnosis/Staging Response Evaluation Surveillance	Quick Superior lung detail Multiplanar reformation capability Usually does not require sedation	Relatively expensive Requires radiation
MR	Diagnosis/Staging Response Evaluation Surveillance	Superior soft-tissue contrast resolution Can provide physiologic information	Long imaging acquisition times May require sedation Limited in evaluating lung parenchyma Relatively expensive
PET/CT	Staging Response Evaluation	Demonstrates metabolically active disease Provides physiologic information May help in biopsy targeting	Expensive Requires radiation May be difficult to discern reactive/inflammatory lymphadenopathy from metastatic disease

CT = computed tomography; MR = Magnetic resonance imaging; PET = positron emission tomograph

Imaging plays a vital role in detecting pediatric primary pulmonary malignancies and associated metastatic disease. CXR is often the study in which primary pulmonary malignancies are demonstrated due to the non-specific respiratory signs and symptoms. We recommend the CXR be obtained with frontal and lateral views to assist in both identifying and localizing the lesion in the airway, lung parenchyma, mediastinum or as multicompartmental. (GRADE D, SOR 1.67, strong recommendation). Although the CXR is typically abnormal in primary pediatric pulmonary tumors, the identification of such lesions may be limited, as tumors can be obscured by associated pneumonia or atelectasis. Furthermore, even though endobronchial neoplasms may be detected on CXR as an abrupt termination of the aerated bronchus with a focal opacity corresponding to a soft tissue mass, small endobronchial lesions may not be apparent. Small pulmonary metastases are often occult on a CXR.

Chest CT is the reference standard for diagnosing and local-regional staging of pediatric pulmonary malignancies due to its speed, high-spatial resolution, and superior depiction of the lung parenchyma compared to other modalities. Thus, we recommend a chest CT as part of the initial evaluation of a primary pediatric pulmonary malignancy. (GRADE A, SOR 1, very strong recommendation)

Chest CT protocols (Table 2) typically include generation of axial slices in soft tissue and lung reconstruction kernels and coronal and sagittal multiplanar reformations. Although the specificity in distinguishing benign and malignant pulmonary nodules is low [5], we strongly suggest CT as the standard for the detection of pulmonary metastases from pediatric solid tumors due to its high sensitivity and the anatomic detail provided for planning of possible metastasectomy. ⁵. (GRADE A, SOR 1.0, very strong recommendation) In the evaluation of metastatic disease, axial slices in lung kernels reconstructed at 1-3 mm thickness are advised, and maximum intensity projections (MIPs) in either the axial or coronal plane should be considered to increase the conspicuity of small nodules that can be missed on standard axial images ⁶. 3D reformations can be generated from the axial data set and assist in biopsy and surgical planning ⁷. 3D-printed models may be created for surgical planning and parental education⁸. Intravenous administration of iodinated contrast agent is generally indicated for assessment of primary pulmonary masses, unless a contraindication to it exists, such as renal insufficiency or severe allergy [11]. Contrast administration is generally not needed if the indication is to assess for pulmonary metastases in the setting of an extrathoracic primary malignancy in which intrathoracic lymphadenopathy is not a concern. CT imaging of the entire chest in cooperative patients can be acquired with a single breathhold in a few seconds or even subsecond scan time with some scanners. For uncooperative or very young children, sedation or anesthesia may be necessary, and alveolar recruitment maneuvers or controlled-ventilation techniques should be considered to minimize atelectasis that can obscure small pulmonary nodules ⁹.

Currently, PPB is the only primary pediatric pulmonary malignancy with imaging staging recommendations issued by a pediatric cancer cooperative group. These were elaborated by the European Cooperative Study Group for Pediatric Rare Tumors (EXPeRT) according to the Concensus Conference Standard Operating Procedure methodology. All patients with PPB should undergo a chest CT with intravenous contrast administration to facilitate assessment of the the lymph nodes, mediastinum, pericardium and diaphragm for possible tumor extension [13]. We concur with these recommendations for PPB. (Grade D, SOR 2, moderate recommendation)

For non-PPB pediatric chest tumors, if the initial CXR reveals a mediastinal process, particularly in the posterior compartment, magnetic resonance imaging (MR) can replace CT as first line choice for initial staging. (GRADE A, SOR 1.33, very strong recommendation) MR scan protocols are highly variable across institutions related to differences in available equipment, expertise, and patient factors and beyond the scope of this article. MR is superior to CT for soft tissue contrast resolution and better delineates mass invasion of the neural foramina and spinal cord. However, MR is of limited value for pulmonary parenchymal involvement given the lack of protons within aerated alveoli, a necessary feature for generating MR signal, and small pulmonary nodules, especially calcified nodules, such as in the setting of metastatic osteosarcoma, may be undetectable [14]. Also, the effectiveness of MR compared to CT in detecting pulmonary nodules may be further compromised by atelectasis related to the longer scan time and more frequent need for sedation with the former. Therefore, we recommend acquiring a dedicated chest CT scan rather than MR if detection of pulmonary metastases is the objective . (GRADE A, SOR 1, very strong recommendation)

Complications, such as post-obstructive pneumonia, can be seen on MR. However, MR requires a longer examination time than CT and possibly sedation ¹⁰, and inadequate/non-diagnostic imaging if the lesion is too small, obscured by pneumonia or if concealed by motion artifact. Thus, we recommend MR serve as a problem-solving tool or in the work-up of possible spinal canal or chest wall invasion. (GRADE B, SOR 1, very strong recommendation)

Inflammatory myofibroblastic tumor (IMT) is a borderline or low-grade malignancy that most commonly occurs in the chest as a pulmonary mass, and less commonly as an endobronchial mass, and can exhibit invasion of the mediastinum, chest wall or diaphragm. Distant metastases are uncommon, occurring in < 5% of cases. IMT has a strong tendency to recur if surgical resection is incomplete, and accurate delineation of disease extent by chest CT or MRis imperative [16].

Although ¹⁸F- fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography (PET)/CT has become the standard of care for assessing thoracic malignancies, such as chest wall rhabdomyosarcoma, Ewing sarcoma, and mediastinal lymphoma, its role in pediatric pulmonary malignanices is not established, being of limited accuracy for nodules less than 5 mm diameter and for discriminating inflammatory processes from aggressive neoplasms ¹¹.

The ligand DOTA-DPhel-Tyr3-octreotate linked to ⁶⁸Gallium serves as a radiotracer for PET/CT or PET/MRI and is highly specific for tumors such as bronchial carcinoids that express somatostatin receptors, providing greater target-to-background ratio and improved lesion conspicuity [18].

EXTRA-THORACIC IMAGING FOR STAGING:

When a pediatric pulmonary malignancy is associated with a high incidence of metastatic disease, a thorough diagnostic imaging search of the most likely involved sites should be undertaken. If the patient has type II or III PPB, they should undergo abdominal CT alongside a brain MR and radionuclide bone scans, as recommend by the EXPeRT group [13]. (Grade D, SOR 2, moderate recommendation)

Abdominal CT can be accomplished by extending the field of view at the time of the initial chest CT examination if the likely nature of the tumor is anticipated in advance. If the patient is cooperative so that sedation or anesthesia is not needed and there is sufficient capacity in the local department, a combination of chest CT and abdominal MR are the best options in order to reduce ionizing radiation exposure and to increase the soft tissue resolution. In a large, multicenter study evaluating whole-body MR for detecting metastates (mainly liver and bone marrow) in the most common pediatric neoplasms, the authors could not establish MR to be equivalent or superior to to conventional imaging paradigms ¹². A subsequent study using diffusion-weighted imaging showed high sensitivity and specificity for the primary tumor and bone and lymph node involvement but low sensitivity for lung metastases in the staging of a spectrum of pediatric tumours ¹³. However, neither this study or other studies have established the superiority of MR over CT for detection of extracranial metastases from pediatric pulmonary malignancies.

¹⁸F-FDG PET/CT is well-established in the primary evaluation and staging of certain adult and pediatric malignancies and is superior to CT alone for the assessment of bone marrow involvement [17]. ¹⁸F-FDG PET/MRI also demonstrates high sensivity for bone marrow infiltration[21]. We recommend ¹⁸F-FDG PET/CT or PET/MRI to be considered as an alternative to radionuclide bone scintigraphy in select cases. (GRADE D, SOR 1.67, very strong recommendation)

Metastases in pediatric bronchial carcinoids occur in 10-15% of cases, most commonly to the liver, adrenal glands, brain, and bones. For the staging of carcinoids, we recommend utilizing somatostatin receptor radiotracers for PET/CT or PET/MR where available ¹⁴. (Grade D, SOR 2.33, moderate recommendation)

TREATMENT RESPONSE EVALUATION:

Typically, post-treatment follow-up imaging comprises the same modalities performed at staging. Unlike more common pediatric malignancies, such as neuroblastoma and lymphoma, no dedicated response criteria have been specifically developed for pediatric primary pulmonary malignancies. Rather, response assessment in some studies rely on more generic established criteria, e.g. Response Evaluation Criteria in Solid Tumors - RECIST 1.1.

SURVEILLANCE IMAGING:

Evidence-based imaging surveillance strategies for detection of relapsed pediatric pulmonary malignancies have not been established. Imaging surveillance strategies using CXR and CT to screen for development of PPB have been proposed for individuals carrying predisposing DICER1 pathogenic gene variants [22]. The use of whole-body MR for screening for malignancies that may involve the chest is discussed in a separate article in this edition.

CONCLUSION:

Evidence-based recommendations are lacking for the imaging of pediatric pulmonary malignancies due to their rarity, and expert opinion forms the basis of the recommendations. Imaging largely relies on chest CT, supplemented with PET/CT or MR to evaluate local-regional extension or metastatic disease in certain settings.

Abbreviation Key:

CT

Computed tomography

MR

Magnetic resonance imaging

CXR

Chest radiograph

PPB

Pleuropulmonary blastoma

IMT

Inflammatory myofibroblastic tumor

¹⁸F-FDG

¹⁸F- fluorodeoxyglucose

SOR

Strength of recommendation