Pediatric Radiology

, Volume 42, Issue 3, pp 331–336

Altered FDG uptake patterns in pediatric lymphoblastic lymphoma patients receiving induction chemotherapy that includes very high dose corticosteroids

Authors

    • Department of Radiology, MLC 5031Cincinnati Children’s Hospital Medical Center
  • Michael J. Gelfand
    • Department of Radiology, MLC 5031Cincinnati Children’s Hospital Medical Center
  • Michael J. Absalon
    • Division of Hematology/OncologyCincinnati Children’s Hospital Medical Center
Original Article

DOI: 10.1007/s00247-011-2228-7

Cite this article as:
Sharp, S.E., Gelfand, M.J. & Absalon, M.J. Pediatr Radiol (2012) 42: 331. doi:10.1007/s00247-011-2228-7

Abstract

Background

Altered FDG uptake patterns were noted in certain lymphoblastic lymphoma patients during therapy.

Objective

To describe these altered FDG uptake patterns and their relationship to chemotherapy.

Materials and methods

Thirty-five FDG PET or PET/CT scans obtained in 11 children with lymphoblastic lymphoma were retrospectively reviewed. FDG uptake patterns were recorded. SUV measurements were performed in liver and facial soft tissues. Results were correlated with induction chemotherapy regimens.

Results

Six of the children had transiently altered FDG uptake with increased uptake in the superficial soft tissues, most notably involving the face. Altered uptake was noted approximately 1 month after initiation of chemotherapy and subsequently resolved. Hepatic uptake was transiently reduced on the 1-month scan in all six children with increased facial uptake. No significant FDG uptake in lymphoma was seen on five of six scans with altered uptake; however, two of these five affected children had FDG uptake in lymphoma on the next follow-up examination. Blood glucose levels in the affected children were in the normal range. All six children with altered FDG uptake received the same induction chemotherapy regimen, which included very high doses of corticosteroids.

Conclusions

Children with lymphoblastic lymphoma on induction chemotherapy protocols including very high doses of corticosteroids transiently demonstrated altered FDG uptake patterns, including increased superficial facial uptake and reduced hepatic uptake. The facial uptake is probably the FDG PET equivalent of Cushingoid facies. Caution in interpreting scans with this altered FDG uptake pattern is suggested, as uptake at sites of lymphomatous involvement may potentially be affected.

Keywords

FDG PETFDG PET/CTLymphoblasticLymphomaCorticosteroidsChildren

Introduction

FDG PET and PET/CT are widely used in lymphoma cases to stage disease and assess response to therapy. While the role of PET imaging after one or two courses of chemotherapy has shown prognostic value in adults with lymphoma, the value of interim PET scans is still under investigation in children [1, 2]. Therapeutic regimens can affect FDG uptake, potentially contributing to misinterpretation of interim evaluations. Marked bone marrow and splenic uptake can be seen after administration of granulocyte colony-stimulating factor [35]. FDG uptake in lymphoma may be temporarily underestimated during and shortly after chemotherapy [6]. In this report, we describe an altered FDG uptake pattern in children with lymphoblastic lymphoma receiving induction chemotherapy including very high doses of corticosteroid therapy.

Materials and methods

Images and patient clinical data were reviewed with permission from the institutional IRB.

PET and PET/CT scans obtained on children with lymphoma during a 6-year period were retrospectively reviewed. Eleven patients with lymphoblastic lymphoma underwent 35 FDG PET or PET/CT scans during the review period. Patients ranged from 2 to 15 years of age at diagnosis. Patients had one to seven scans during the review period.

FDG PET scans were performed 40 to 60 min after the administration of 5.2 MBq/kg (0.14 mCi/kg) of FDG. When PET/CT was performed, the CT dose was reduced to 35% of the beam current normally used for a diagnostic CT at our institution. All PET and PET/CT scans were performed when ordered for clinical indications by the referring physicians. In some children, timing of FDG PET was mandated by multicenter chemotherapy protocols.

Imaging findings were reviewed by a pediatric nuclear medicine physician and a pediatric radiologist with nuclear medicine interest. The pattern of FDG uptake in the head and neck was recorded. SUV measurements were made in liver and soft tissues of the face in children with altered FDG uptake. SUV measurements could not be obtained in one patient due to incompatibility of header information between an earlier PET camera and the current PET workstation.

The chemotherapy regimen and phase of therapy were noted and correlated with timing of the PET scans.

Results

Eleven children with lymphoblastic lymphoma underwent 35 FDG PET or PET/CT scans during the review period. Nine of these children had FDG PET or PET/CT scans performed during the first 3 months of therapy (range 8–89 days). One child was scanned only at diagnosis and one child had an initial post-therapy scan 103 days after initiation of chemotherapy.

Six children had transiently altered FDG uptake on scans obtained approximately 1 month (27–32 days) after initiation of chemotherapy with increased uptake in the skin and subcutaneous fat, most notably involving the face (Figs. 1, 2 and 3). Increased FDG uptake in the face was unrelated to expected normal salivary gland or muscle activity, localizing to the subcutaneous fat of prominent cheeks/jowls. In five children with increased FDG uptake in whom SUV could be measured, maximum SUV in superficial facial tissues was 4.7, 3.5, 2.3, 1.7, and 1.3 respectively. Variably increased FDG uptake was also seen in the skin and subcutaneous fat of other areas, including the shoulders, trunk and posterior neck (Figs. 1, 2 and 3); this subcutaneous fat uptake was not in the distribution of brown fat.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-011-2228-7/MediaObjects/247_2011_2228_Fig1_HTML.gif
Fig. 1

A 14-year-old with lymphoblastic lymphoma imaged 1 month into therapy. a MIP viewed from anterior projection. b–d Coronal PET/CT images of the face. e–g Axial PET/CT images of the face. Diffusely increased uptake is noted in the skin/subcutaneous tissues, most notably affecting the cheeks/jowls, which are prominent (maximum SUV 3.5). Hepatic FDG uptake is mildly reduced from baseline (mean SUV 2.0)

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-011-2228-7/MediaObjects/247_2011_2228_Fig2_HTML.gif
Fig. 2

A 4-year-old with lymphoblastic lymphoma imaged 1 month into therapy. a MIP viewed from anterior projection. b–d Axial PET/CT images of the face. e–g Axial PET/CT images of the abdomen. Markedly increased uptake is noted diffusely in the skin/subcutaneous tissues (facial max SUV 1.3). Hepatic FDG uptake is markedly reduced (mean SUV 0.5)

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-011-2228-7/MediaObjects/247_2011_2228_Fig3_HTML.gif
Fig. 3

A 9-year-old with lymphoblastic lymphoma imaged 1 month into therapy (also shown in Fig. 4). a MIP viewed from anterior projection. b–c Sagittal PET/CT images of the face/neck. d–e Axial PET/CT images of the face. f–g Axial PET/CT images of the abdomen. Increased uptake is noted within the skin/subcutaneous tissues of the cheeks/jowls, which are prominent (maximum SUV 4.7). Increased uptake is also noted in the posterior soft tissues of the neck (“buffalo hump” indicated by arrows). Hepatic FDG uptake is reduced (mean SUV 1.0). Diffuse hepatic steatosis is noted

Hepatic FDG uptake was reduced from the level seen on baseline and/or subsequent scans in all six children with altered uptake patterns. In five children in whom SUV could be measured, mean SUV of the liver was 1.0, 2.0, 0.4, 1.2 and 0.5, respectively. Localization CT revealed diffuse hepatic steatosis in two affected children (Fig. 3). Hepatic steatosis resolved on follow-up imaging.

In five of six scans with altered FDG uptake patterns, no significant FDG uptake was seen at previously detected sites of lymphoma. However, two of these five affected children had FDG uptake in lymphoma on follow-up examinations performed 6 weeks later (Figs. 4 and 5). These two children had recurrent uptake in mediastinal masses that also increased in size on the follow-up localization CT scans. In the other three children, the mediastinal masses all deceased in size on the next CT examination with no recurrence of FDG avidity.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-011-2228-7/MediaObjects/247_2011_2228_Fig4_HTML.gif
Fig. 4

A 9-year-old with lymphoblastic lymphoma (also shown in Fig. 3). a–c Axial PET/CT images obtained at diagnosis show marked uptake within a very large mediastinal mass (maximum SUV 18.1). The child was imaged in decubitus position as lying supine induced respiratory distress. d–f Axial PET/CT images obtained 1 month into therapy show only small foci of low-level uptake, not definitely corresponding to the residual mediastinal mass. g–i Axial PET/CT images obtained 6 weeks later show FDG uptake in the mediastinal mass, which has slightly increased in size (maximum SUV 3.6)

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-011-2228-7/MediaObjects/247_2011_2228_Fig5_HTML.gif
Fig. 5

A 5-year-old with lymphoblastic lymphoma. a Chest X-ray performed at diagnosis shows a large mediastinal mass. The child was unable to undergo cross-sectional imaging due to severe respiratory distress in non-upright positions. b–d Axial PET/CT images performed 1 month into therapy show increased uptake in the subcutaneous tissues and axillary fat without uptake in the residual mediastinal mass. e–g Axial PET/CT images performed 6 weeks later show FDG uptake in the residual mediastinal mass, which has increased in size (maximum SUV 3.3)

One scan with altered FDG uptake showed persistent uptake within the medullary canal of the right femur at a site of presumed lymphomatous involvement; the primary facial mass had markedly decreased in size and demonstrated no FDG uptake.

Blood glucose levels in the affected children were in the normal range (72 – 98 mg/dL).

The altered FDG uptake pattern was only seen in children imaged approximately 1 month (27–32 days) after initiation of induction chemotherapy with resolution on follow-up scans. All six children with altered FDG uptake patterns had received the same induction chemotherapy protocol with vincristine, daunorubicin, L-asparaginase and very high doses of prednisone (60 mg/m2 per day for 28 days followed by a 10-day taper). This chemotherapy protocol is used in both Europe and North America.

Three other children received the same induction chemotherapy protocol, but were not imaged at the 1-month point; one child was scanned 8 days after initiation of induction chemotherapy, while the second was scanned 16 and 41 days after induction of chemotherapy, and the third was scanned 74 days after induction of chemotherapy. Their scans did not exhibit the altered FDG uptake patterns.

Two additional children did not exhibit altered FDG uptake. One was scanned only at diagnosis. The second received induction chemotherapy with vincristine, doxorubicin, methotrexate, 6-mercaptopurine and prednisone (40 mg/m2 per day for 21 days) and was imaged 103 days after initiation of induction chemotherapy.

Chemotherapy protocols and scan timing are summarized in Table 1.
Table 1

Summary of protocols and scan timing

Protocola

Patient

Scan timing (days from start of induction)

Altered FDG uptake

Unaltered FDG uptake

1

1

31

−1, 80

1

2

27

−4, 74, 109, 136

1

3

30

77, 133, 161, 197, 253, 386

1

4

31

−4, 17, 87

1

5

32

−1, 88, 841

1

6

31

−16, −2

1

7

 

8

1

8

 

−6, 16, 41

1

9

 

74

2

10

 

−9

2

11

 

−2, 103, 187

aInduction chemotherapy protocols:

Protocol 1 = Vincristine (days 0, 7, 14, 21), daunorubicin (days 0, 7, 14, 21), L-asparaginase (day 11, then 3 doses/week for 9 doses) and prednisone (60 mg/m2 per day for 28 days followed by a 10-day taper)

Protocol 2 = Vincristine (days 1, 8, 15, 22), doxorubicin (days 1, 2, 22), methotrexate (day 2), 6-mercaptopurine (days 22–35), prednisone (40 mg/m2 per day for 21 days; no taper)

Discussion

We describe a transiently altered FDG uptake pattern including increased superficial facial and decreased hepatic uptake in certain children with lymphoblastic lymphoma undergoing induction chemotherapy. All affected children received the same induction chemotherapy protocol, including prednisone 60 mg/m2 per day for 28 days followed by a 10-day taper. The altered FDG uptake pattern occurred only on scans obtained approximately 1 month after initiation of the chemotherapy regimen. Children scanned at other times did not exhibit altered FDG uptake patterns.

Corticosteroid use over extended time periods (greater than 3–4 weeks) can result in Cushing syndrome. Characteristic features of Cushing syndrome include moon facies, supraclavicular fat distribution, dorsocervical “buffalo hump” and truncal obesity.

The altered FDG uptake pattern noted in our study follows the expected distribution of Cushingoid characteristics, most notably Cushingoid moon facies. A few children also had increased uptake in prominent fat in the posterior neck, in the expected location of a Cushingoid “buffalo hump.” Increased subcutaneous fat uptake was also sometimes seen in the shoulders/supraclavicular regions and trunk.

The fact that the altered FDG uptake pattern was noted only at the 1-month point correlates well with the timing of corticosteroid administration during induction chemotherapy. Children imaged earlier during induction chemotherapy were receiving corticosteroids, but had not yet had time to develop Cushingoid characteristics. Children imaged laters had already completed their corticosteroid tapers.

Hepatic steatosis has been reported in patients receiving chemotherapy [7] and corticosteroids [811] and was transiently seen in two affected patients. The reduced hepatic uptake seen in our study is likely unrelated to hepatic steatosis, as hepatic steatosis has previously been shown to have no significant effect on hepatic FDG uptake in adults [12].

The effect of this altered FDG uptake pattern on FDG avidity at sites of lymphoma involvement is unclear. Five of six scans with altered FDG uptake patterns showed no significant uptake in lymphoma, although follow-up examinations showed FDG-avid mediastinal masses in two of these five children. Given that these mediastinal masses also enlarged on the follow-up localization CT scans, the lack of FDG uptake could represent near-complete eradication of the tumor with subsequent regrowth. However, suppression of tumor uptake by the very high corticosteroid doses is a possible alternative explanation. Caution in the interpretation of FDG PET scans may therefore be warranted if these altered uptake patterns are noted.

Hyperglycemia during FDG PET scanning has previously been shown to decrease detection rates of some cancers [1315], while showing no effect on sensitivity in other studies [16, 17]. Although all of our patients had normal glucose levels at the time of their studies, glucose metabolism was likely impacted by the very high doses of corticosteroids.

The altered FDG uptake pattern noted in our study has not been seen in any other patient group we have studied. Previous publications describing FDG PET or PET/CT scans in Cushingoid patients with ACTH-secreting tumors or adrenal hyperplasia have not reported altered FDG uptake patterns [1826].

Our study is limited by the lack of a control population and the small number of children. The focused disease population in our study also limits broader generalization.

Conclusion

Children with lymphoblastic lymphoma on induction chemotherapy protocols that include very high doses of corticosteroids transiently demonstrated altered FDG uptake patterns, including increased superficial facial uptake and reduced hepatic uptake. The facial uptake is probably the FDG PET equivalent of Cushingoid facies. Caution in interpreting scans with this altered FDG uptake pattern is suggested, as FDG uptake at sites of lymphomatous involvement may potentially be affected.

Copyright information

© Springer-Verlag 2011