AJR 2001; 176:755-759
© American Roentgen Ray Society
Pulmonary Metastases at Diagnosis of Neuroblastoma in Pediatric Patients
CT Findings and Prognosis
Bamidele F. Kammen1,2,
Katherine K. Matthay3,
Preeyacha Pacharn1,4,
Robert Gerbing5,
Robert C. Brasch1,3 and
Charles A. Gooding1,3
1
Department of Radiology, University of California San Francisco, 505 Parnassus
Ave., San Francisco, CA 94143.
2
Present address: Department of Diagnostic Imaging, Children's Hospital
Oakland, 747 Fifty-Second St., Oakland, CA 94609.
3
Department of Pediatrics, University of California San Francisco, San
Francisco, CA 94143.
4
Present address: Department of Radiology, Mahidol University, Bangkok,
Thailand.
5
The Children's Oncology Group, 440 Hustingon Dr., Arcadia, CA 91066.
Received June 30, 2000;
accepted after revision August 28, 2000.
Supported by grant CA13539 (Children's Cancer Group) from the Division of
Cancer Treatment, National Cancer Institute, National Institutes of Health,
Department of Health and Human Services; and the Tkalcevic and Kasle
Neuroblastoma Research Fund, the Conner Fund, and the Campini Fund.
Address correspondence to C. A. Gooding.
Abstract
OBJECTIVE. We undertook this study to determine the frequency, CT
appearance, and clinical implications of the rare occurrence of pulmonary
metastases among children presenting with neuroblastoma.
MATERIALS AND METHODS. A search of the Children's Cancer Group
database revealed 21 of 567 children with reported lung metastases at original
diagnosis of neuroblastoma. CT examinations available for 17 of these patients
were analyzed retrospectively to determine if lung metastases were present,
and if so, to characterize their radiographic features.
RESULTS. Seventeen (3%) of 567 patients presenting with Evans stage
IV neuroblastoma had confirmed pulmonary metastases at diagnosis. All had
metastases to at least one site other than the lungs. The most common CT
appearance of pulmonary lesions was of up to five, small, bilateral,
noncalcified nodules. In nine patients (53%), the pulmonary nodules initially
resolved with treatment. In this cohort, six children developed progressive
disease and died, and three are still alive. All eight children whose lung
lesion did not completely respond to treatment died. Overall, children with
pulmonary metastases had unfavorable Shimada histology, a higher association
with amplification of the MYCN oncogene (p = 0.0002), and a decreased
event-free survival (p < 0.001) when compared with all children
with stage IV neuroblastoma without pulmonary metastases.
CONCLUSION. The search for neuroblastoma lung metastases, which
occur more frequently than previously reported, is clinically important
because their presence portends a poor prognosis.
Introduction
Neuroblastoma is the most common extracranial solid tumor in children,
accounting for 10% of all childhood cancers
[1,
2]. Most children presenting
with neuroblastoma have metastatic disease at diagnosis; however, their
prognosis varies according to many clinical and biologic risk factors
[2,3,4,5,6,7].
Neuroblastoma lung metastases, particularly at presentation, are rare, and the
lungs have been reported to be an unfavorable metastatic site
[3]. To our knowledge, the CT
appearance of lung metastases in children presenting with neuroblastoma has
not been previously reported. We therefore performed this study to determine
the incidence and characteristic radiographic findings of neuroblastoma
pulmonary metastases.
Materials and Methods
A search of the records of all patients with Evans stage IV neuroblastoma
[4] registered on the
Children's Cancer Group (CCG) protocols 3881 (open June 1989 to August 1996)
and 3891 (open January 1991 to April 1996) found 21 patients with reported
lung metastases at diagnosis
[3]. Lung metastasis was
defined as tumor in the lung parenchyma
[3]. CT examinations for 17 of
21 patients with pulmonary metastases were obtained, and these patients (10
boys and 7 girls) with a median age of 2.1 years (age range, 4 months to 5.2
years) constituted our study population. Radiologic examinations for the
remaining four children were not available. Institutional review board
approval was obtained for protocols CCG 3881 and CCG 3891 as was informed
consent for each patient.
At diagnosis, all patients underwent contrast-enhanced chest and abdominal
CT performed with helical or axial technique with a slice thickness ranging
from 5 to 10 mm, according to the standard protocols of the respective
institutions caring for the patient. Initial unenhanced images were available
for 12 cases. All studies were performed with window and level settings
appropriate for visualization of soft tissue and lung parenchyma. Four images
per patient were obtained for bone detail as well. Follow-up chest CT
examinations were available for 16 of the 17 patients.
All studies were analyzed independently by three experienced pediatric
radiologists with attention to the number, size, configuration, and
distribution of the pulmonary metastases. Other metastatic sites and the
location and extent of the primary were recorded. Follow-up chest CT
examinations were reviewed to determine the evolution of lung metastases.
Comparison of the biologic features of the tumor (serum ferritin level
[5], MYCN oncogene
amplification [6], Shimada
histopathology [7]) and the
event-free survival between the subgroup of patients with pulmonary metastases
and all other children with stage IV disease was made by the chisquare test of
proportions.
Results
CT findings confirmed that 17 (3%) of 567 children had lung metastases at
diagnosis of neuroblastoma. Histologic confirmation was available in one
patient. Of the 17 patients with pulmonary metastases for whom CT examinations
were obtained, 15 (88%) had an adrenal primary tumor compared with 69% of all
patients with stage IV neuroblastoma. Two of the 17 children had a nonadrenal
primary tumor, one in the pelvis and one in the thorax. All patients with lung
metastases had metastatic disease in at least one other site, which in order
of decreasing frequency were bone marrow (n = 11), bone (n =
10), and liver (n = 6). The Shimada histopathologic classification
was unfavorable in all patients tested, and the ferritin level was elevated in
44% (Table 1). The MYCN
oncogene was amplified in all children with lung metastases and in 34% of all
other children with stage IV disease (p = 0.0002)
(Table 1).
View this table:
[in this window]
[in a new window]
|
TABLE 1 Unfavorable Biologic Features of Tumor in Children With and Without
Neuroblastoma Lung Metastases at Diagnosis
|
|
Lung metastases typically included multiple nodules (median number, five
nodules) with most nodules between 0.5 and 1 cm in diameter (Figs.
1,2,3A,3B).
In two children, lung nodules were as large as 3 cm. Nodules were randomly
distributed throughout the lungs with a slightly increased predilection for
the lung bases and showed smooth and spiculated margins (Fig.
4A,4B).
Except for one child with a thoracic primary tumor who also had a single
ipsilateral lung metastasis, all other children had at least two lesions.
Calcification was not identified in any of the nodules.
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —2-year-old boy at diagnosis of neuroblastoma.
Contrast-enhanced CT scan of abdomen shows right adrenal primary
neuroblastoma. Note liver and vertebral metastases. Lesion abuts but does not
invade kidney.
|
|
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —2-year-old boy at diagnosis of neuroblastoma. CT scan of lung
shows characteristic appearance of neuroblastoma lung metastases. Note
bilateral nodules that are as large as 1 cm in diameter.
|
|
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —2-year-old girl with neuroblastoma. Chest CT scans show
multiple pulmonary metastatic nodules ranging in diameter from 0.3 cm to 2.5
cm. Margin of nodules may be smooth (straight arrow, A) or
spiculated (curved arrow, B).
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —2-year-old girl with neuroblastoma. Chest CT scans show
multiple pulmonary metastatic nodules ranging in diameter from 0.3 cm to 2.5
cm. Margin of nodules may be smooth (straight arrow, A) or
spiculated (curved arrow, B).
|
|
Occasionally, pulmonary parenchymal metastases were associated with
mediastinal or pleural disease, or both. Mediastinal lymphadenopathy was
identified in six children, and pleural thickening was identified in seven
children with pleural-based nodules; two children developed pleural effusions
during the course of their disease.
Lung metastases resolved during initial treatment for nine patients. In
this cohort, three patients had recurrent disease in the lung and died of
their disease (Fig.
5A,5B,5C).
Of the six patients who did not have recurrent lung metastases, three had
progression of disease in other sites and died, and three are still alive
(range, 3.6-5.5 years; mean survival to date, 4.7 years). Among the three
children still living, two are disease-free and one has residual disease in
the abdomen. All eight children who did not have a complete response to
therapy died of their disease. Overall, 14 of the 17 patients died.
View larger version (93K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C. —Neuroblastoma lung metastases in 4-month-old boy. CT scan
obtained 10 months after B shows recurrence of pulmonary metastatic
nodules, most occurring at same sites as in A. Patient died of
overwhelming disease.
|
|
Discussion
Neuroblastoma lung metastases are widely believed to be a terminal event,
occurring only when tumor is widely disseminated
[8,9,10,11].
In our analysis, pulmonary involvement at initial diagnosis of neuroblastoma
occurs in at least 3% of cases. This number is probably an underestimate,
because CT examinations were not available for all patients with reported lung
metastases. Moreover, because we could not review chest studies on all
children with stage IV neuroblastoma, additional cases may have been missed.
Regardless, we calculate the prevalence of neuroblastoma lung metastases at
diagnosis to be higher than values previously reported
[12].
Cowie et al. [12]
calculated the incidence of pulmonary involvement at neuroblastoma diagnosis
to be 0.7% by reviewing data on 1245 children in the European Neuroblastoma
Study Group. In their study, treating institutions provided information on the
sites of metastatic disease and lung lesions that were visible on radiographs
or on CT [12]. Of note, their
definition of pulmonary metastases was broader than ours and included children
with pleural disease [12]; if,
as we did, they had included children with only parenchymal disease, their
calculated frequency of pulmonary involvement would have been even less: four
(0.3%) of 1245. The discrepancy between the higher incidence of pulmonary
metastases in our study and the value found by Cowie et al. may be due to the
fact that all children in our study had CT examinations available that enabled
us to detect small lesions that could be missed on radiographs.
Other reports of neuroblastoma pulmonary metastatic disease include
children with an established diagnosis of neuroblastoma and children who were
inadvertently reinjected with malignant cells for bone marrow transplantation
[8,9,10,11,12].
Towbin and Gruppo [8] reported
radiographic patterns of neuroblastoma pulmonary lesions (lymphangitic
metastases, hematogenous spread, and direct extension) in children with widely
disseminated disease.
We did not identify any cases with interstitial engorgement or
bronchovascular enlargement that, if present, would suggest lymphangitic
spread of tumor. The studies that we reviewed were routine chest CT
examinations with the slice thickness ranging from 6 to 10 mm; however, with
this technique interstitial disease can be missed. High-resolution CT images
are much more sensitive for the detection of lymphangitic disease. In the
future, survey images using this technique could be obtained at the time of
chest evaluation.
In our study, the CT appearance of neuroblastoma pulmonary metastases of
multiple, bilateral nodules, randomly dispersed throughout the lung with a
slight basilar predominance, most likely represented hematogenous
distribution. In one patient, only a single lesion was identified on the
ipsilateral side as a thoracic primary lesion, which was compressing the right
upper lobe bronchus. This may have represented bronchogenic spread of
disease.
The characteristic nodules should easily be seen on conventional
radiographs, but because smaller lesions would be difficult to detect, a
routine chest CT at diagnosis is probably warranted. Interestingly, none of
the pulmonary nodules in our patients were calcified, despite the fact that
calcification is identified on CT in 85% of primary neuroblastoma lesions
[13].
Our study lacks pathologic confirmation in all but one case, because the
diagnosis of distant metastases was made at biopsy of an alternative
nonpulmonary site and subsequent biopsy of the lung nodules would not have
changed clinical treatment. This raises the question of accuracy when
diagnosing the lung as a metastatic site. Radiographically, these lesions
looked like metastases (smooth-bordered or spiculated nodules); moreover, the
nodules resolved after chemotherapy in nine of the children (Fig.
5A,5B,5C).
Most of these patients relapsed in the lung, and similar lesions reappeared in
many of the same sites, supporting the fact that these were indeed metastatic
lesions and not infection (Fig.
5A,5B,5C).
The remaining eight children never completely responded to treatment, and
their lung disease progressed. In the future, scintigraphic imaging with
metaiodobenzyguanidine would be a potentially useful, noninvasive way to
characterize and follow up pulmonary lesions.
Biologic features of neuroblastoma have been previously shown to be
unfavorable in children with neuroblastoma lung metastases, with MYCN
amplification and unfavorable Shimada histopathology in nearly all children
tested [3]
(Table 1). Moreover, the
event-free survival is much lower for children with neuroblastoma pulmonary
metastases as compared with all other children with stage IV disease, 15%
versus 50%, respectively, two years after diagnosis (p < 0.001)
[3]. It seems unlikely that the
presence of lung metastases actually causes a decreased event-free survival.
Instead, the presence of pulmonary metastases probably reflects more
biologically aggressive disease as reflected by the poor tumor biology.
In conclusion, neuroblastoma pulmonary metastases are more common than
previously reported. Although the incidence of lung involvement at diagnosis
of neuroblastoma is only 3%, the inclusion of the chest when the initial
abdominal CT is obtained is warranted because the presence of lung metastases
portends a poor prognosis and most of these patients die of their disease.
Acknowledgments
We thank the contributing Children's Cancer Group investigators and
institutions: R. Hutchinson (University of Michigan Medical Center, Ann Arbor,
MI; CA02971), K. Matthay (University of California Medical Center, San
Francisco, CA; CA17829), D. Puccetti (University of Wisconsin Hospital,
Madison, WI; CA05436), J. R. Geyer (Children's Hospital and Medical Center,
Seattle, WA; CA10382), E. Kodish (Rainbow Babies and Children's Hospital,
Cleveland, OH; CA20320), G. Reaman (Children's National Medical Center,
Washington, DC; CA03888), P. Gaynon (Children's Hospital of Los Angeles, Los
Angeles, CA; CA02649), J. Whitlock (Vanderbilt University School of Medicine,
Nashville, TN; CA26270), J. Neglia (University of Minnesota Health Sciences
Center, Minneapolis, MI; CA07306), W. Carroll (University of Utah Medical
Center, Salt Lake City, UT; CA10198), P. Rogers (University of British
Columbia Vancouver, B. C., Canada; CA29013), R. Wells (Children's Hospital
Medical Center, Cincinnati, OH; CA26126), S. Feig (University of California
Medical Center, Los Angeles, CA; CA27678), C. Arndt (Mayo Clinic and
Foundation, Rochester, MN; CA 28882), R. Drachtman (University of Medicine and
Dentistry of New Jersey, Camden, NJ), M. Hetherington (Children's Mercy
Hospital, Kansas City, MO), B. Raney (M. D. Anderson Cancer Center, Houston,
TX), and V. Shen (Children's Hospital of Orange County, Orange, CA).
We thank Wendy Neale for assistance with manuscript preparation.
References
-
Young JL, Ries LG, Silverberg E, Horm JW, Miller RW. Cancer
incidence, survival and mortality for children younger than age 15 years.
Cancer
1986;58:598
-602[Medline]
-
Matthay KK, Villablanca JG, Seeger RC, et al. Treatment of
high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous
bone marrow transplantation and 13-cis-retinoic acid. N
Engl J Med 1999;341:1165
-1173[Abstract/Free Full Text]
-
DuBois SG, Kalika Y, Lukens JN, et al. Metastatic sites in stage IV
and IVS neuroblastoma correlate with age, tumor biology and survival.
J Pediatr Hematol Oncol
1999;21:181
-189[Medline]
-
Evans AE. Staging and treatment of neuroblastoma.
Cancer
1980;45:1799
-1802[Medline]
-
Hann HWL, Evans AE, Siegel SE, et al. Prognostic importance of
serum ferritin in patients with stages III and IV neuroblastoma: the
Children's Cancer Study Group experience. Cancer Res
1985;45:2843
-2848[Abstract/Free Full Text]
-
Seeger RC, Brodeur GM, Sather H, et al. Association of multiple
copies of the N-myc oncogene with rapid progression of neuroblastomas.
N Engl J Med
1985;313:1111
-1116[Abstract]
-
Shimada H, Chatten J, Newton WA, et al. Histopathologic prognostic
factors in neuroblastic tumors: definition of subtypes of ganglioneuroblastoma
and age-linked classification of neuroblastomas. J Natl Cancer
Inst 1984;73:405
-416
-
Towbin R, Gruppo RA. Pulmonary metastases in neuroblastoma.
AJR
1982;138:75
-78[Abstract/Free Full Text]
-
Glorieux P, Bouffet E, Philip I, et al. Metastatic interstitial
pneumonitis after autologous bone marrow transplantation: a consequence of
reinjection of malignant cells? Cancer
1986;58:2136
-2139[Medline]
-
Graeve JLA, De Alarcon PA, Sato Y, Pringle K, Helson L. Miliary
pulmonary neuroblastoma: a risk of autologous bone marrow transplantation?
Cancer
1988;62:2125
-2127[Medline]
-
Kalra R, Zoger S, Kosovich MS, Matthay KK. Radiological case of the
month. Arch Pediatr Adolesc Med
1995;149:194
-195
-
Cowie F, Corbett R, Pinkerton CR. Lung involvement in
neuroblastoma: incidence and characteristics. Med Pediatr
Oncol 1997;28:429
-432[Medline]
-
Barnewolt CE, Paltiel HJ, Lebowitz RL, Kirks DR. Genitourinary
tract. In: Kirks DR, Griscom NT, eds. Practical pediatric imaging:
diagnostic radiology of infants and children. Philadelphia:
Lippincott-Raven, 1998:1010
-1170
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
