[18F]fluorodeoxyglucose Positron Emission Tomography for

VOLUME 29 䡠 NUMBER 26 䡠 SEPTEMBER 10 2011
JOURNAL OF CLINICAL ONCOLOGY
O R I G I N A L
R E P O R T
[18F]Fluorodeoxyglucose Positron Emission Tomography
for Detection of Bone Marrow Involvement in Children and
Adolescents With Hodgkin’s Lymphoma
Sandra Purz, Christine Mauz-Körholz, Dieter Körholz, Dirk Hasenclever, Antje Krausse, Ina Sorge,
Kathrin Ruschke, Martina Stiefel, Holger Amthauer, Otmar Schober, W. Tilman Kranert, Wolfgang A. Weber,
Uwe Haberkorn, Patrick Hundsdörfer, Karoline Ehlert, Martina Becker, Jochen Rössler, Andreas E. Kulozik,
Osama Sabri, and Regine Kluge
Sandra Purz, Dirk Hasenclever, Antje
Krausse, Ina Sorge, Osama Sabri, and
Regine Kluge, University of Leipzig,
Leipzig; Christine Mauz-Körholz, Dieter
Körholz, Kathrin Ruschke, and Martina
Stiefel, Martin-Luther-University HalleWittenberg, Halle/Saale; Holger
Amthauer and Patrick Hundsdörfer,
Charité University Medicine Berlin,
Berlin; Otmar Schober, University
Hospital Münster; Karoline Ehlert,
University Children’s Hospital Münster,
Münster; W. Tilman Kranert and
Martina Becker, Johann Wolfgang
Goethe University, Frankfurt; Wolfgang
A. Weber, University of Freiburg;
Jochen Rössler, University Medical
Hospital Freiburg, Freiburg; Uwe Haberkorn, University of Heidelberg; and
Andreas E. Kulozik, Children’s Hospital,
University of Heidelberg, Heidelberg,
Germany.
Submitted September 2, 2010;
accepted May 5, 2011; published online
ahead of print at www.jco.org on
August 8, 2011.
Supported by grants from the Peter
Escher Foundation for Children With
Cancer, Menschen für Kinder, Hand in
Hand for Children, the Lions
Kinderkrebs-Forschungs-und Ausbildungszentrum, Grit Jordan Foundation,
and Deutsche Krebshilfe (M.S.).
Both S.P. and C.M-K. contributed
equally to this work.
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this
article.
Corresponding author: Sandra Purz,
MD, Department of Nuclear Medicine,
University of Leipzig, Liebigstrasse 18,
04103 Leipzig, Germany; e-mail:
[email protected].
© 2011 by American Society of Clinical
Oncology
0732-183X/11/2926-3523/$20.00
DOI: 10.1200/JCO.2010.32.4996
A
B
S
T
R
A
C
T
Purpose
Currently, a routine bone marrow biopsy (BMB) is performed to detect bone marrow (BM)
involvement in pediatric Hodgkin’s lymphoma (HL) stage greater than IIA. [18F]fluorodeoxyglucose
positron emission tomography (FDG-PET) is increasingly used for the initial staging of HL. The
value of using FDG-PET to detect BM involvement has not been sufficiently defined. We
compared the results of BMBs and FDG-PET for the diagnosis of BM involvement in a large
pediatric group with HL.
Patients and Methods
The initial staging of 175 pediatric patients with newly diagnosed classical HL stage greater than
IIA was determined by using BMB, FDG-PET, chest computed tomography (CT), and magnetic
resonance imaging (MRI) or CT of the neck, abdomen, and pelvis. Staging images were
prospectively evaluated by a central review board. Skeletal regions that were suggestive of BM
involvement by either method were re-evaluated by using different imaging modalities. In
suspicious cases, bone scintigraphy was performed. If follow-up FDG-PET scans were available,
the remission of skeletal lesions during treatment was evaluated.
Results
BMB results were positive in seven of 175 patients and were identified by FDG-PET. FDG-PET
scans showed BM involvement in 45 patients. In addition, the lesions of 32 of these 45 patients
had a typical multifocal pattern. In 38 of 39 follow-up positron emission tomography scans, most
of the skeletal lesions disappeared after chemotherapy. There was no patient with skeletal
findings suggestive of BM involvement by MRI or CT with a negative FDG-PET.
Conclusion
FDG-PET is a sensitive and specific method for the detection of BM involvement in pediatric HL.
The sensitivity of a BMB appears compromised by the focal pattern of BM involvement. Thus,
FDG-PET may safely be substituted for a BMB in routine staging procedures.
J Clin Oncol 29:3523-3528. © 2011 by American Society of Clinical Oncology
INTRODUCTION
The staging of Hodgkin’s lymphoma (HL) has
evolved with the refinement of computed tomography (CT) and magnetic resonance imaging (MRI)
methods and the introduction of [18F]fluorodeoxyglucose positron emission tomography (FDG-PET).
Step-by-step invasive staging procedures have been
abandoned. Although the majority of patients in the
1970s underwent a laparotomy,1 this invasive diagnostic procedure has been almost completely abandoned. A bone marrow biopsy (BMB) for the
detection of bone marrow (BM) involvement re-
mains the only invasive staging procedure. However, the detection of BM involvement is important
because it implicates stage IV disease and necessitates the stratification to the most intense treatment group.
BM involvement was detected in 4.8% to 14% of
adults with HL.2-4 Several studies with small numbers
of patients indicated a limited sensitivity of a BMB and
a higher sensitivity of FDG-PET to detect BM involvement in malignant lymphoma.5,6 For children and adolescents with HL, only limited data are available.
We addressed the issue of whether the use of
FDG-PET negates the need for a BMB in HL staging
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
3523
Purz et al
Table 1. Definition of Tumor Stages and TGs
TG*†
No. of
Chemotherapy
Cycles
Radiation at
the End of
Chemotherapy?
1, early stages
2
IEA/B, IIEA, IIB,
2, intermediate
or IIIA
stages
IIEB, IIIEA/B, IIIB, 3, advanced
or IVA/B
stages
4
Yes, if no CR
after chemotherapy
Yes
6
Yes
Ann Arbor
Stage
IA/B and IIA
Abbreviations: CR, complete remission; TG, treatment group (see MauzKörholz et al7).
ⴱ
According to the Gesellschaft für Pädiatrische Onkologie und Hämatologie Hodgkin’s Disease 2002 (GPOH-HD2002) trial and the GPOH-HD2002 Vinblastine, Etoposide, Cyclophosphamide, Vincristine, Prednisolone, Adriamycine trial.
†Only patients in TG2 and 3 were included in the present analysis.
procedures. We also sought to clarify whether the use of FDG-PET
instead of a BMB resulted in a relevant change in tumor staging and
the use of more intensive treatment.
PATIENTS AND METHODS
Patients
Between 2002 and 2006, a total of 676 children with newly diagnosed HL
were enrolled in the Gesellschaft für Pädiatrische Onkologie und Hämatologie
Hodgkin’s Disease 2002 (GPOH-HD2002)7 or GPOH-HD2002 Vinblastine, Etoposide, Cyclophosphamide, Vincristine, Prednisolone, Adriamycine (VECOPA)
treatment optimization studies. Both trials were approved by the Ethics Committee of the University of Leipzig and the respective institutional review boards of
participating centers. All patients or guardians of patients gave written informed
consent to participate in the trials. Patients were assigned to one of three treatment
groups (Table 1) according to their Ann Arbor stage. According to the study
protocols, the initial staging included chest CT scans, MRI or CT scans of the neck,
abdomen, and pelvis, and BMBs in patients with HL greater than stage IIA. When
skeletal involvement was suspected, bone scintigraphy (BS) was recommended.
Positron emission tomography (PET) scanning was not mandatory but was routinely performed in several participating centers. Inclusion and exclusion criteria
for the analysis are listed in Table 2.
Imaging
FDG-PET. FDG-PET was performed by applying the standard procedures for children. These were defined within the recommendations of the
pediatric task group of the European Association of Nuclear Medicine.8 In
Table 2. Inclusion and Exclusion Criteria
Inclusion Criteria
Exclusion Criteria
Age ⬍ 18 years
First-line treatment according to GPOHHD2002 or GPOH-HD2002-VECOPA
protocol
Stage ⬎ IIA
Pretreatment FDG-PET
Pretreatment BMB
First FDG-PET after start of
chemotherapy
Lymphocyte-predominant HL
Unknown BMB result
Abbreviations: BMB, bone marrow biopsy; FDG-PET, 关18F兴fluorodeoxyglucose positron emission tomography; GPOH-HD2002, Gesellschaft für Pädiatrische Onkologie
und Hämatologie Hodgkin’s Disease 2002 trial; GPOH-HD2002-VECOPA, GPOHHD2002 Vinblastine, Etoposide, Cyclophosphamide, Vincristine, Prednisolone, Adriamycine trial; HL, Hodgkin’s lymphoma.
3524
Table 3. Staging Imaging Performed in Cohort of 175 Patients
Imaging Modality
Imaging Location
No. of Patients
CT
MRI
CT
MRI
CT
MRI
CT
MRI
BS
Neck
Neck
Chest
Chest
Abdomen
Abdomen
Pelvis
Pelvis
Whole body
91
95
171
51
86
106
85
106
73
Abbreviations: BS, bone scintigraphy; CT, computed tomography; MRI,
magnetic resonance imaging.
detail, dedicated PET or PET/CT scanners (Biograph Duo, ECAT EXACT 922,
ECAT HR⫹ [Siemens Healthcare Germany, Erlangen, Germany]; Allegro
[Philips Healthcare Europe, Best, The Netherlands]; Discovery LS [GE Healthcare Germany, Munich, Germany]) were used. After a fasting period of 4 to 6
hours, an acquisition of the attenuation-corrected scan was started 40 to 90
minutes after the intravenous administration of a body weight–adapted dose
of [18F]fluorodeoxyglucose.9 The examination included the entire region
from the skull to upper thighs. If BM involvement was suspected, additional
imaging of the lower extremities was recommended.
Because PET scanning was not mandatory in this trial, it was not routinely performed in all participating centers. Therefore, to test for a selection
bias, we compared the demographic characteristics of both groups of patients
with or without PET.
CT/MRI. CTwastobeperformedinallregionsfromtheepipharynxtothe
lower edge of the pelvic symphysis, at a layer thickness of 5 mm, with the administration of oral contrast medium and the body weight–adapted administration of
intravenous contrast (1.5 to 2.0 mL/kg). Alternatively, if MRI of neck, abdominal,
and pelvic sites was performed, it included T2-weighted, fat-saturated, transverse,
and coronal sequences and T1-weighted dynamic sequences with a contrast agent
in the arterial, portal venous, and venous phases.
BS. Whole-body BS was only recommended in patients with pathologic
skeletal findings by BMB, PET, or CT/MRI. A body surface–adapted dosing of
99m
Tc-diphosphonates according to guidelines for BS in children10 was advised.
The number of patients with MRI or CT scans and BS results at staging is listed in
Table 3.
BMB
BMBs were performed in standard regions at the iliac crest. Written
reports of BMB results were sent to the central study tumor board for review.
Central Review of Imaging
All staging images (PET, MRI, CT, and BS) were reviewed at the tumor
board at the start of the treatment of each patient.
In the first step, two reference nuclear physicians evaluated the PET
without knowing the radiology results. In a separate setting, the reference
radiologist reviewed the CT/MRI without knowing the PET results. At the
weekly interdisciplinary reference tumor board, the nuclear physician presented the BM PET results, and thereafter, the radiologist systematically
checked all positive areas for correlative findings.
On PET scans, the BM and associated bony involvement was diagnosed
when there was an enhancement of FDG uptake in single or multiple foci in the
skeleton (Fig 1). This pattern was differentiated from a diffuse, enhanced
uptake in the skeleton, which is frequently seen in patients with HL and
considered to be paraneoplastic bone-marrow activation. A quantitative analysis of FDG uptake was not performed.
On MRI scans, the BM involvement was diagnosed if focal lesions were
seen in T2- or T1-weighted sequences or both.
If PET-positive lesions outside routine MRI/CT regions were detected at
the time of the central review, no additional correlative MRI was performed
because the central review usually took place after the start of treatment.
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
JOURNAL OF CLINICAL ONCOLOGY
FDG-PET Diagnostic in Pediatric Hodgkin’s Lymphoma
A
B
C
Fig 1. Thirteen-year-old girl with a negative bone marrow biopsy (patient identification number 1202, Table 4; images courtesy of University of Aachen, Aachen,
Germany). (A and B) Positron emission tomography shows focal enhanced [18F]fluorodeoxyglucose uptake (arrows) in multiple areas of the thoracic spine, sternum,
and pelvic bones. (C) Computed tomography scan shows corresponding lytic lesions (arrows) in both iliac crests.
Skeletal lesions that were typical for other skeletal diseases, such as bone
cysts or lesions caused by a BMB, were documented to be negative in respect to
skeletal HL and were not further evaluated.
caused by other etiologies such as a benign bone tumor, trauma, or inflammation (ie, they represented a low level of confirmation).
Reference Levels for BM Involvement Detected by FDG-PET
Ideally, the calculation of a positive predictive value (PPV) would require
the histologic examination of PET-positive lesions. To calculate the negative
predictive value (NPV), sensitivity and specificity because of the focal pattern
of BM involvement in the histologic examination of HL in the entire central
skeleton would be the gold standard. However, for practical reasons, skeletal
biopsies could not be performed in addition to standard BMBs in the pediatric
patient group. Therefore, the NPV and sensitivity were calculated by using the
results of BMBs and MRIs, CTs, or both as surrogate markers. Consequently,
if no skeletal abnormality suggestive of HL was detected with any other
method, a negative FDG-PET scan was considered to be a true negative.
Standard definitions of diagnostic test characteristics (ie, PPV, NPV, true
positive, false positive, true negative, and false negative) were used. The PPV was
calculated as the number of true positives (test-positive patients who fulfilled the
reference standard) divided by the number of test positives (patients in whom the
imaging method showed a positive finding). The NPV was calculated as the
number of true negatives (test-negative patients who did not have the disease)
divided by the number of test negatives (patients in whom the imaging method
was negative).
TocalculatePPVandspecificity,acombinationoftheresultsoftheBMB,the
pattern of PET findings, and the PET response to treatment were used. Therefore
true positivity and false-negativity of PET results were defined by three reference
levels (RLs) as follows: a positive BMB (level a, the highest level of confirmation); a
multifocal pattern of skeletal PET findings with three or more lesions in the central
skeleton, that is, vertebrae, pelvis, chest, and proximal parts of humeri or femora
(level b, a high level of confirmation because this pattern was typical for HL and
rarely occurs in other skeletal diseases11); and a complete or almost complete
resolution in follow-up PET scans performed during or at the end of chemotherapy (level c, an intermediate level of confirmation because the disappearance of
inflammatory or traumatic lesions might also occur).
The detection of a corresponding lesion on the MRI, CT, or BS was not
considered to be RL because these corresponding lesions might also have been
RESULTS
www.jco.org
A total of 404 of 676 patients presented with classical HL stage greater
than IIA. Staging FDG-PET scanning was performed in 199 of 404
patients before the initiation of treatment. In 24 of 199 patients, no
BMB was performed, or no information about the result of the BMB
was available. Therefore, 175 patients (mean age, 14.6 years; 89 boys
and 86 girls) from 39 centers fulfilled all inclusion criteria and had no
exclusion criteria. In contrast, 229 patients with classical HL greater
than stage IIA either did not fulfill the inclusion criteria or had an
exclusion criterion. Neither group differed in sex, extranodal involvement, or “B” symptoms. Patients who did not fulfill the inclusion
criteria were slightly younger (mean age, 13.9 v 14.6 years; data
not shown).
Role of FDG-PET for Detection of BM Involvement
Compared With BMB and Morphologic Imaging
In 175 patients with HL stage greater than IIA, the disease of 45
patients (25.7%) was PET positive for BM involvement (Table 4). PET
results were questionable in three of 175 patients and negative in 127 of
175 patients.
Only seven of 175 patients (4%) had positive BMB results. All
seven of these patients had at least one FDG-PET–positive lesion in the
skeleton. In all 28 patients with a positive finding by MRI or CT, at
least one correlative lesion was FDG-PET positive. Thus, no falsenegative PET lesions were found. Therefore, the sensitivity and NPV
of PET were 100% when using a combination of BMB, CT, or MRI
results for reference.
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
3525
Purz et al
Table 4. Findings in Biopsy and Imaging, Staging, and Treatment Groups of 45 Patients With Possible Bone Marrow Involvement
Patient
ID
BMB
PET
MRI
CT
001
037
1004
1044
1046
1056
1086
1098
1181
1193
1202
1214
1218
122
1220
1225
1226
1256
131
1810-9
259
295
251
1067
1804-0
1243
029
051
1028
104
1095
1210
1238
1807-1
1811-2
247
260
299
1106‡
1152‡
1223‡
1232‡
1269‡
210‡
282‡
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
NA
Pos
Pos
NA
NA
Pos
Pos
Pos
Pos
NA
Pos
NA
Pos
Pos
Pos
Pos
NA
NA
NA
NA
Pos
NA
NA
NA
NA
NA
NA
NA
NA
NA
Neg
NA
Pos
Pos
Pos
Pos
Pos
NA
Neg
NA
NA
NA
Neg
NA
NA
Pos
Pos
NA
Pos
Pos
Neg
Pos
NA
NA
Pos
Pos
Pos
Neg
NA
NA
NA
Quest
NA
NA
NA
NA
Neg
Neg
Neg
NA
Neg
NA
Neg
Neg
NA
NA
Neg
Pos
Neg
NA
Quest
Pos
NA
NA
BS
PET
Response
Number of
Skeletal Lesions
in FDG-PET
NA
Pos
Pos
Pos
NA
Neg
Pos
Neg
NA
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Pos
Neg
Pos
Neg
Pos
Pos
Pos
Neg
Pos
Pos
Neg
Neg
Neg
Pos
NA
Neg
NA
Neg
NA
Neg
NA
Pos
Neg
Pos
Pos
NA
Neg
NA
Adeq
Adeq
Adeq
Adeq
NA
NA
Adeq
Adeq
NA
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
NA
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
NA
Adeq
Adeq
Adeq
Adeq
Adeq
NA
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Adeq
Unclear
Adeq
Adeq
3
⬎3
⬎3
⬎3
⬎3
⬎3
⬎3
⬎3
1
⬎3
⬎3
3
⬎3
⬎3
2
1
⬎3
⬎3
⬎3
3
1
⬎3
1
1
3
⬎3
⬎3
⬎3
⬎3
⬎3
⬎3
1
1
3
1
⬎3
1
1
⬎3
⬎3
⬎3
⬎3
1
⬎3
1
RLⴱ
b,c
b,c
b,c
b,c
b
b
b,c
b,c
b,c
b,c
b,c
b,c
b,c
c
c
b,c
b
b,c
b,c
c
b,c
c
c
b,c
b
b,c
b,c
b,c
b,c
b,c
c
b,c
c
b,c
c
c
a,b,c
a,b,c
a,b,c
a,b,c
a
a,b,c
a,c
Change of Stage
Because of Skeletal
PET Finding
Change of TG
Because of Skeletal
PET Finding†
⫹1
⫹2
⫹2
⫹1
⫹1
⫹2
⫹2
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹2
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹1
⫹2
⫹1
Abbreviations: Adeq, adequate (complete disappearance of the majority of skeletal lesions in [18F]fluorodeoxyglucose positron emission tomography [FDG-PET]);
BMB, bone marrow biopsy; BS, bone scintigraphy; CT, computed tomography; ID, identification number; MRI, magnetic resonance imaging; NA, not available (not
performed/not evaluable); Neg, negative; PET, positron emission tomography; Pos, positive; Quest, questionable; RL, reference level.
ⴱ
Level a, positive BMB; level b, multifocal pattern of skeletal PET findings with three or more lesions in the central skeleton; level c, complete or almost complete
resolution in follow-up PET during or at the end of chemotherapy.
†TG1 comprised stages IA/B and IIA; TG2 comprised stages IEA/B, IIEA, IIB, and IIIA; and TG3 comprised stages IIEB, IIIEA/B, IIIB, and IVA/B.
‡Patients were assigned to stage IV (ie, TG3) because of a positive BMB.
Among 168 BMB-negative patients, FDG-PET results were negative in 127 and questionable in three patients. Of 45 FDG-PET–
positive patients, only seven patients were BMB positive. In 32 of 45
patients, three or more skeletal lesions were present on FDG-PET
scans; 12 patients had only one such lesion, and one patient had two
such lesions. Of these 13 patients, two patients had BMB-positive
3526
disease. In 39 of 45 FDG-PET–positive patients, follow-up PET scans
after two (n ⫽ 16), three (n ⫽ 1), four (n ⫽ 8), or six (n ⫽ 14) courses
of chemotherapy were available. In 38 of 39 patients, the majority of
skeletal lesions that were detectable by FDG-PET disappeared completely after chemotherapy. Specificity and PPV for the 3 RLs are listed
in Table 5.
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
JOURNAL OF CLINICAL ONCOLOGY
FDG-PET Diagnostic in Pediatric Hodgkin’s Lymphoma
Table 5. Specificity and PPV of FDG-PET for Diagnosis of Bone Marrow
Involvement by Use of Different Reference Levels
Specificity
PPV
Levelⴱ
No.
%
No.
%
a
b
c
130 of 168
130 of 141
130 of 132
77.3
92.2
98.5
7 of 45
34 of 45
43 of 45
15.6
75.5
95.6
Abbreviations: BMB, bone marrow biopsy; FDG-PET, 关18F兴fluorodeoxyglucose positron emission tomography; PPV, positive predictive value.
ⴱ
Level a, BMB; level b, BMB or multifocality (⬎ two lesions in central
skeleton); level c, BMB or multifocality or response to chemotherapy (complete disappearance of PET findings in the majority of skeletal lesions).
In 25 of 45 PET-positive patients, MRI results were available in at
least one of the PET-positive skeletal areas and were positive in 23 or 25
FDG-PET–positive patients (92%). Of 45 PET-positive patients, 40 patients were evaluated for bone involvement by using BS (n ⫽ 36) and/or
CT (n ⫽ 23). In 25 of 40 patients (62.5%), at least one of the PET–positive
lesions showed a correlative finding on BS (23 of 36) and/or CT (10 of 23)
scans (Table 4). In most patients, BS scans detected more lesions than did
CT. Because of the limited confirmative value of such corresponding
lesions, these data were not used to calculate the PPV and specificity.
Change of Tumor Stage and Treatment Group When
Using FDG-PET Instead of BMB for Detection of
BM Involvement
A comparison of the difference in event-free survival between the
130 PET-negative and 45 PET-positive patients suggested a trend
toward worse prognoses in PET-positive patients; however, because of
the overall low rate of events, the results were not statistically significant (Appendix Fig A1, online only).
As listed in Table 4, 38 of 45 patients had negative BM aspirates;
seven of 45 patents had BM aspirates. In the group of patients with
negative BM aspirates, FDG-PET findings would have led to a change
in disease stage in 20 of 38 patients. In 10 of 38 patients, PET results
and a change in stage would also have affected to which treatment
group the patient was assigned. Seven of 10 patients had intensified
treatment on the basis of additional correlative imaging results. There
were no relapses in three of 10 patients with PET-positive lesions who
did not receive intensified therapy. In the group of patients with
positive BM aspirates, all seven patients were determined to have stage
IV disease and were treated accordingly. In these patients, there were
no false-negative PET scans, and the concordant PET and BM results
suggested that PET may be safely substituted for BM-aspirate results
without negatively affecting the stage or treatment group assignment.
If only the typical multifocal pattern of BM involvement had
been accepted for the diagnosis of BM involvement, 13 of 168 patients
(7.7%) would have been upstaged, and treatment would have been
intensified in five patients (2.9%). However, one of the seven BMBpositive patients would have been downstaged (Table 4; patient identification number 282) without changing the treatment intensity. A
second patient (patient identification number 1269) was diagnosed
with HL and showed only a single bony lesion in the pelvis and an
adjacent soft tissue mass without nodal involvement. HL was confirmed after a targeted biopsy of this lesion. In addition, seven of 32
patients with a typical multifocal pattern of BM involvement relapsed,
but none of the 13 patients with fewer than three lesions relapsed.
www.jco.org
DISCUSSION
Currently, histologic samples of the BM are usually obtained from the
iliac crest and are considered to be the gold standard for the evaluation
of BM involvement in pediatric HL. This method is based on the
assumption that, in cases of BM involvement, tumor cells spread
through the marrow and may, therefore, be detected by a focal biopsy,
although sampling errors are recognized. However, with the introduction of FDG-PET into the staging procedures of HL, it became evident
that the typical involvement of BM may not be homogenous but,
rather, focal. Therefore, a routine iliac crest BMB can miss the focal
involvement and, thus, carries a low sensitivity.11,12
Our results from a large pediatric group with advanced HL confirmed the occurrence of a focal involvement pattern of the BM, a low
sensitivity of the BMB but a high sensitivity and NPV, and an acceptable specificity and PPV of FDG-PET.
BM involvement is rare in pediatric patients with HL. In this
study, BM involvement was diagnosed by using BMBs in only 4% of
children and adolescents with classical HL stage greater than IIA. This
rate was consistent with other studies on pediatric HL,13-15 which
showed positive BMB results in up to 6.5% of patients. These rates
were markedly lower than those in adults, in whom 4.8% to 14% of
BMBs were positive.3,11,16-24
FDG-PET– detected BM-positivity rates are much higher than
BMB-detected BM-positivity rates. In our study, diagnoses were determined for 45 of 175 patients by PET versus seven of 175 by BMB.
Similarly, another pediatric group reported nine of 25 patients diagnosed by PET and zero of 25 patients by BMB.15 However, our study
included only intermediate and high stages because no BMB was
performed in early-stage disease. In the GPOH-HD-2002 trial of 101
early-stage patients, 100 patients had a completely negative PET scan
in the skeleton (ie, only one patient showed a PET-positive lesion
without a morphologic correlate; data not shown). Therefore, a
consideration of all HL-stages would result in a much lower rate of
PET-positive skeletal findings. In adults, FDG-PET–positive but
BMB-negative findings occurred in 5% to 13% of patients.6,11,12
FDG-PET–positive lesions were often correlated with positive
results by using other imaging techniques (23 of 25 positive results by
MRI, 23 of 36 positive results by BS, and 10 of 23 positive results by
CT). In our study, all 28 findings by conventional imaging were
concomitantly FDG-PET positive, which showed the superior sensitivity of FDG-PET compared with that of other imaging methods.
A definitive validation of true positivity in PET-positive lesions
would require local biopsies. Because of practical constraints, this was
not possible in our pediatric patients. Moreover, only limited data are
available in the literature. From the available data of several authors,5,6,25 12 patients were identified. In all of these patients, a tumor
involvement of FDG-PET–positive lesions was histologically confirmed, which suggested a high specificity of PET, although a publication bias could not be excluded.
FDG-PET–positive lesions in BM often display a multifocal pattern.5,6,11,18,26,27 In our study, 32 of 45 patients presented with three or
more lesions in the central part of the skeleton, including vertebrae,
pelvis, chest, or proximal ends of humeri or femora. This typical
pattern is easier to diagnose in a pediatric population because the
skeletal activity distribution is not compromised by degenerative disorders, which are often seen in adult patients.
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
3527
Purz et al
FDG-PET–positive lesions in HL typically respond to currently recommended chemotherapy programs.11,12 In 38 of 39 follow-up PETs in
our study, BM lesions responded during or at the end of chemotherapy.
Therefore,anadequateresponsetochemotherapyisanadditionalparameter to aid in the assessment of FDG-PET–positive BM lesions.
In conclusion, our study confirmed previous observations that
described the pattern of BM involvement in HL as focal and typically
multifocal. A focal involvement pattern explains why untargeted
BMBs sampled from the iliac crest have low sensitivity. In contrast,
FDG-PET has a high sensitivity and specificity. Thus, a routine BMB
could be omitted. In place of a BMB, a typical multifocal pattern of
skeletal PET findings could be considered another means of assessing
BM involvement in patients with HL. In case of solitary skeletal PET
findings, a targeted biopsy should be performed if the diagnosis of BM
involvement would change the treatment recommendation.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
OF INTEREST
Although all authors completed the disclosure declaration, the following
author(s) indicated a financial or other interest that is relevant to the subject
matter under consideration in this article. Certain relationships marked
with a “U” are those for which no compensation was received; those
relationships marked with a “C” were compensated. For a detailed
description of the disclosure categories, or for more information about
REFERENCES
1. Hennekeuser HH, Möbius W, Obrecht P, et
al: Bone marrow biopsy in malignant lymphomas [in
German]. Klin Wochenschr 52:118-122, 1974
2. Munker R, Hasenclever D, Brosteanu O, et al:
Bone marrow involvement in Hodgkin’s disease: An
analysis of 135 consecutive cases. German Hodgkin’s
Lymphoma Study Group. J Clin Oncol 13:403-409, 1995
3. Vassilakopoulos TP, Angelopoulou MK,
Constantinou N, et al: Development and validation
of a clinical prediction rule for bone marrow involvement in patients with Hodgkin lymphoma. Blood
105:1875-1880, 2005
4. O’Carroll DI, McKenna RW, Brunning RD:
Bone marrow manifestations of Hodgkin’s disease.
Cancer 38:1717-1728, 1976
5. Schaefer NG, Strobel K, Taverna C, et al: Bone
involvement in patients with lymphoma: The role of
FDG-PET/CT. Eur J Nucl Med Mol Imaging 34:60-67,
2007
6. Moog F, Bangerter M, Kotzerke J, et al:
18-F-fluorodeoxyglucose-positron emission tomography as a new approach to detect lymphomatous
bone marrow. J Clin Oncol 16:603-609, 1998
7. Mauz-Körholz C, Hasenclever D, Dörffel W, et
al: Procarbazine-free OEPA-COPDAC chemotherapy
in boys and standard OPPA-COPP in girls have
comparable effectiveness in pediatric Hodgkin’s
lymphoma: The GPOH-HD-2002 Study. J Clin Oncol
10:3680-3686, 2010
8. Stauss J, Franzius C, Pfluger T, et al: Guidelines for 18F-FDG PET and PET-CT imaging in paediatric oncology. Eur J Nucl Med Mol Imaging 35:
1581-1588, 2008
9. Lassmann M, Biassoni L, Monsieurs M, et al:
The new EANM paediatric dosage card: Additional
ASCO’s conflict of interest policy, please refer to the Author Disclosure
Declaration and the Disclosures of Potential Conflicts of Interest section in
Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory
Role: None Stock Ownership: None Honoraria: None Research
Funding: Wolfgang A. Weber, Philips Medical Systems, Bayer Expert
Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Sandra Purz, Christine Mauz-Körholz, Dieter
Körholz, Osama Sabri, Regine Kluge
Administrative support: Dieter Körholz, Osama Sabri
Provision of study materials or patients: Christine Mauz-Körholz, Dieter
Körholz, Holger Amthauer, Otmar Schober, Uwe Haberkorn, Patrick
Hundsdörfer, Karoline Ehlert, Martina Becker, Jochen Rössler, Andreas E.
Kulozik, Regine Kluge
Collection and assembly of data: Sandra Purz, Christine Mauz-Körholz, Dieter
Körholz, Antje Krausse, Ina Sorge, Martina Stiefel, Holger Amthauer, W. Tilman
Kranert, Wolfgang A. Weber, Uwe Haberkorn, Patrick Hundsdörfer, Karoline
Ehlert, Martina Becker, Jochen Rössler, Andreas E. Kulozik, Regine Kluge
Data analysis and interpretation: Sandra Purz, Christine Mauz-Körholz,
Dieter Körholz, Dirk Hasenclever, Ina Sorge, Kathrin Ruschke, Otmar
Schober, Karoline Ehlert, Osama Sabri, Regine Kluge
Manuscript writing: All authors
Final approval of manuscript: All authors
notes with respect to F-18. Eur J Nucl Med Mol
Imaging 35:1666-1668, 2008
10. Hahn K, Fischer S, Colarinha P, et al: Guidelines for bone scintigraphy in children. Eur J Nucl
Med 28:42-47, 2001
11. Pelosi E, Penna D, Deandreis D, et al: FDGPET in the detection of bone marrow disease in
Hodgkin’s disease and aggressive non-Hodgkin’s
lymphoma and its impact on clinical management.
Quart J Nucl Med Mol Img 52:9-16, 2008
12. Cerci JJ, Praccia LF, Soares J, et al: Positron
emission tomography with 2-[18F]-fluoro-2-deoxy-Dglucose for initial staging of Hodgkin lymphoma: A single
center experience in Brazil. Clinics (Sao Paulo) 64:491498, 2009
13. Simpson CD, Gao J, Fernandez CV, et al:
Routine bone marrow examination in the initial evaluation of paediatric Hodgkin lymphoma: The Canadian perspective. Br J Haematol 141:820-826, 2008
14. Mahoney DH Jr, Schreuders LC, Gresik MV,
et al: Role of staging bone marrow examination in
children with Hodgkin disease. Med Pediatr Oncol
30:175-177, 1998
15. Shulkin BL, Goodin GS, McCarville MB, et
al: Bone and [18F]fluorodeoxyglucose positronemission tomography/computed tomography scanning for the assessment of osseous involvement in
Hodgkin lymphoma in children and young adults.
Leuk Lymphoma 50:1794-1802, 2009
16. Horan FT: Bone involvement in Hodgkin’s
disease. A survey of 201 cases. Br J Surg 56:277281, 1969
17. Feltl D, Markova J, Mocikova H, et al: Prognostic impact of bone involvement in Hodgkin lymphoma. Neoplasma 55:96-100, 2008
18. Carr R, Barrington SF, Madan B, et al: Detection of
lymphoma in bone marrow by whole-body positron
emission tomography. Blood 91:3340-3346, 1998
19. Rosenberg SA: Hodgkin’s disease of the bone
marrow. Cancer Res 31:1733-1736, 1971
20. Menon NC, Buchanan JG: Bilateral trephine
bone marrow biopsies in Hodgkin’s and nonHodgkin’s lymphoma. Pathology 11:53-57, 1979
21. Doll DC, Ringenberg QS, Anderson SP, et al:
Bone marrow biopsy in the initial staging of Hodgkin’s disease. Med Pediatr Oncol 17:1-5, 1989
22. Bartl R, Frisch B, Burkhardt R, et al: Assessment of bone marrow histology in Hodgkin’s disease: Correlation with clinical factors. Br J Haematol
51:345-360, 1982
23. Ellis ME, Diehl LF, Granger E, et al: Trephine
needle bone marrow biopsy in the initial staging of
Hodgkin disease: Sensitivity and specificity of the
Ann Arbor staging procedure criteria. Am J Hematol
30:115-120, 1989
24. Howell SJ, Grey M, Chang J, et al: The value
of bone marrow examination in the staging of
Hodgkin’s lymphoma: A review of 955 cases seen
in a regional cancer centre. Br J Haematol 119:
408-411, 2002
25. Buchmann I, Reinhardt M, Elsner K, et al:
2-(fluorine-18)fluoro-2-deoxy-D-glucose positron emission tomography in the detection and staging of malignant lymphoma. A bicenter trial. Cancer 91:889-899,
2001
26. Fuster D, Chiang S, Andreadis C, et al: Can
[18F]fluorodeoxyglucose positron emission tomography imaging complement biopsy results from the
iliac crest for the detection of bone marrow involvement in patients with malignant lymphoma? Nucl
Med Commun 27:11-15, 2006
27. Moog F, Kotzerke J, Reske SN: FDG PET
can replace bone scintigraphy in primary staging
of malignant lymphoma. J Nucl Med 40:14071413, 1999
■ ■ ■
3528
© 2011 by American Society of Clinical Oncology
Downloaded from ascopubs.org by 78.47.27.170 on January 13, 2017 from 078.047.027.170
Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
JOURNAL OF CLINICAL ONCOLOGY